METHODE ET DISPOSITIF DE CONTROLE DES PARAMETRES DE PRODUCTION

METHOD OF CONTROLLING PRODUCTION PROCESSES AND APPARATUS THEREFOR

Abrégé anglais

ABSTRACT OF THE INVENTION
A method of controlling a process using a
three-state four-mode process controller including the steps
of providing a desired value signal to said process control-
ler related to the condition to which it is desired to set
said process, providing a feedback signal to said process
controller from the process being controlled indicating the
current condition of the process, providing a reset state
signal to said process controller adapted to insure that the
process controller can be easily reset to a state one condi-
tion as desired, which causes said process controller to begin
operation in its first state producing a correction signal
causing the process device which changes the condition of the
process to operate at a predetermined rapid speed in a first
desired direction, utilizing said desired value, feedback,
and reset state signals to cause said process controller to
change to its second state of operation when the error dif-
ference between said desired value and said feedback signal
changes polarity thereby producing a correction signal caus-
ing said process device to operate at a predetermined rapid
speed in the opposite direction, utilizing said error differ-
ence, the rate of change of said error difference, and the
reset state signal to cause said process controller to change
to its third state of operation when the summation of said
error difference and said rate of change of said error dif-
ference changes polarity, producing a correction signal from
said desired value and said feedback signals while said pro-
cess controller is in its third state which will look ahead

and attempt to become saturated as soon as a new desired
value is supplied or a process change occurs by utilizing
said rate of change and said error difference, which will
remain unchanged as long as the process being controlled, and
said desired value signal both remain unchanged and in a
static condition, which will, if saturated, be brought out of
saturation by utilizing said error difference and said rate
of change in a manner to change said correction signal in
value much faster than if said error difference only were
used, and which will, if said process is in a dynamic condi-
tion, be changed in a series of occurrences to a value smal-
ler in magnitude, but of either polarity, until it arrives at
a value related to said condition it is desired to set said
process to, and utilizing said correction signal to cause
said process to arrive at said desired condition.

Revendications

WE CLAIM:
1. A method of controlling a process using a
three-state four-mode process controller including the steps
of providing a desired value signal to said process control-
ler related to the condition to which it is desired to set
said process, providing a feedback signal to said process
controller from the process being controlled indicating the
current condition of the process, providing a reset state
signal to said process controller adapted to insure that the
process controller can be easily reset to a state one condi-
tion as desired, which causes said process controller to begin
operation in its first state producing a correction signal
causing the process device which changes the condition of the
process to operate at a predetermined rapid speed in a first
desired direction, utilizing said desired value, feedback,
and reset state signals to cause said process controller to
change to its second state of operation when the error dif-
ference between said desired value and said feedback signal
changes polarity thereby producing a correction signal caus-
ing said process device to operate at a predetermined rapid
speed in the opposite direction, utilizing said error differ-
ence, the rate of change of said error difference, and the
reset state signal to cause said process controller to change
to its third state of operation when the summation of said
error difference and said rate of change of said error dif-
ference changes polarity, producing a correction signal from
said desired value and said feedback signals while said pro-
cess controller is in its third state which will look ahead
78

and attempt to become saturated as soon as a new desired
value is supplied or a process change occurs by utilizing
said rate of change and said error difference, which will
remain unchanged as long as the process being controlled, and
said desired value signal both remain unchanged and in a
static condition, which will, if saturated, be brought out of
saturation by utilizing said error difference and said rate
of change in a manner to change said correction signal in
value much faster than if said error difference only were
used, and which will, if said process is in a dynamic condi-
tion, be changed in a series of occurrences to a value smal-
ler in magnitude, but of either polarity, until it arrives at
a value related to said condition it is desired to set said
process to, and utilizing said correction signal to cause
said process to arrive at said desired condition.
2. The method defined in Claim 1, wherein the steps
of providing said desired value signal and of providing said
reset state signal to said process controller include the
steps of providing a desired setting device capable of sup-
plying a reference voltage signal and said reset state signal,
connecting said desired setting device in an appropriate
manner to said process controller, and setting said desired
setting device such that said desired reference voltage signal
will be an output therefrom and said reset state signal will
be a second output therefrom.
3. The method defined in Claim 2, wherein the step
of providing said desired setting device and connecting said
79

desired setting device include the steps of providing a
potentiometer and a pushbutton switch and directly connecting
said potentiometer and said pushbutton switch to said process
controller.
4. The method defined in Claim 1, wherein the step
of providing said desired value signal and said reset state
signal to said process controller includes the steps of pro-
viding a desired setting device capable of supplying a refer-
ence voltage signal and said reset state signal, connecting
to said desired setting device an automation device to auto-
matically select said reset state signal if desired and auto-
matically change said reference voltage signal from said
desired setting device to a value appropriate to a next con-
dition upon the completion of a test, and connecting said
desired setting device to said process controller.
5. The method defined in Claim 1, wherein the step
of providing said feedback signal to said process controller
includes the steps of providing a process measurement device
capable of measuring the current state of the process being
controlled, causing said process measurement device to supply
a process correlate signal related to the current condition
of the process, causing said process correlate signal to
either be directly supplied to said process controller or to
a feedback signal device capable of converting or signal
conditioning said process correlate signal into a signal
usable by and directly supplied to said process controller

such that said signal supplied to said process controller is
said feedback signal related to the current condition of the
process being controlled.
6. The method defined in Claim 1, wherein the step
of producing a correction signal by utilizing said desired
value, feedback and reset state signals includes the steps of
providing a three state error and rate amplifier circuit, sup-
plying said three state error and rate amplifier circuit with
said desired value, feedback and reset state signals, and
yielding a correction signal related to said reset state signal
and to the algebraic sum of the actual error difference and the
rate of change of said actual error difference between said
feedback and desired value signals.
7. The method defined in Claim 1, wherein the step
of producing a correction signal by utilizing said desired
value feedback, and reset state signals includes the steps of
providing a three state error and rate amplifier circuit,
providing a valid range check circuit, supplying said desired
value, feedback, and reset state signals to said three state
error and rate amplifier circuit, supplying high limit and low
limit set points and said desired value signal into said valid
range check circuit, providing an output from said valid range
check circuit adapted to produce a saturation override signal
if the desired value is outside said high or low limit set
points, supplying said saturation override signal to said
three state error and rate amplifier circuit, causing said
81

three state error and rate amplifier circuit to provide a cor-
rection signal which is saturated when said process controller
is in its first state or its second state or when said desired
value signal is either above said high limit set point or below
said low limit set point, and causing said three state error
and rate amplifier circuit to provide a correction signal
related to the algebraic sum of the actual error difference
and the rate of change of said actual error difference between
said feedback and desired value signals.
8. The method defined in Claim 7, wherein the step
of providing said three state error and rate amplifier circuit
includes the steps of providing a first operational amplifier
having positive and negative inputs and an output, providing a
second operational amplifier having positive and negative
inputs and an output, providing an instrumentation amplifier
having positive and negative inputs, an output, and gain set
inputs, providing an edge detector having an input, a pulse
output, and a ground, providing a state counter device having
a reset state input, a clock input, a clock inhibit input, a
state one output, a state two output, and a state three output,
connecting said reset state signal to said reset state input
of said state counter device, connecting a polarity signal
corresponding to the polarity of said correction signal to
said input of said edge detector connecting said saturation
override signal to said negative input of said instrumentation
amplifier, connecting said output of said edge detector to
said clock input of said state counter device, connecting said
82

ground of said edge detector to ground, connecting said
desired value signal to said positive input of said first oper-
ational amplifier, connecting said feedback signal to said
positive input of said second operational amplifier, interpos-
ing a capacitor between said negative inputs of said first and
said second operational amplifiers, connecting said state
three output of said state counter device to said clock inhibit
input thereof, connecting a first state three, a second state
three and a third state three analog switch to said state
three output of said state counter device, connecting a first
state two, a second state two and a third state two analog
switch to said state two output of said state counter device,
connecting a first state one, a second state one and a third
state one analog switch to said state one output of said state
counter device, connecting said output of said first opera-
tional amplifier to said positive input of said instrumentation
amplifier, connecting said first state three analog switch,
said first state two analog switch, and said first state one
analog switch to the negative input of said first operational
amplifier, connecting between the output of said first opera-
tional amplifier and said first state three analog switch a
first state three variable resistor, also connecting between
the output of said first operational amplifier and said first
state two analog switch a first state two variable resistor,
connecting between the output of said first operational ampli-
fier and said first state one analog switch a first state one
variable resistor, connecting said second state three analog
switch, said second state two analog switch, and said second
83

state one analog switch to the negative input of said second
operational amplifier, connecting between the output of said
second operational amplifier and second state three analog
switch a second state three variable resistor, connecting
between the output of said second operational amplifier and
said second state two analog switch a second state two
variable resistor, connecting between the output of said
second operational amplifier and said second state one analog
switch a second state one variable resistor, connecting across
said gain set inputs of said instrumentation amplifier in
series a third state three variable resistor and said third
state three analog switch) connecting across said gain set
inputs of said instrumentation amplifier in series a third
state two variable resistor and a third state two analog
switch, connecting across the gain set input of said instrumen-
tation amplifier in series a third state one variable resistor
and said third state one analog switch, interposing a resistor
between said output of said second operational amplifier and
said negative input of said instrumentation amplifier, and
obtaining said correction signal from said output of said
instrumentation amplifier and providing a state one signal
from said state one output of said state counter device.
9. The method defined in Claim 8, wherein the step
of producing a correction signal causing the process device
which changes the condition of the process to operate at a
predetermined rapid speed in a first desired direction
includes the steps of causing said state counter device to
84

operate in its state one condition thereby connecting said
state one output to said first state one analog switch, said
second state one analog switch, and said third state one
analog switch thereby causing said first state one variable
resistor to be connected in series with said first state one
analog switch and causing both of said devices to be connected
between said output of said first operational amplifier and
said negative input of said first operational amplifier, caus-
ing said second state one variable resistor to be connected in
series with said second state one analog switch and causing
both of said devices to be connected between said output of
said second operational amplifier and said negative input of
said second operational amplifier, causing said third state
one variable resistor to be connected in series with said
third state one analog switch and causing both of said devices
to be connected between said gain set inputs of said instrumen-
tation amplifier, thereby obtaining said correction signal from
said output of said instrumentation amplifier which causes
said process device to operate at a predetermined rapid speed
in a first desired direction.
10. The method defined in Claim 9, wherein the step
of producing a correction signal causing said process device
to operate at a predetermined rapid speed in the opposite
direction includes the steps of causing said state counter
device to operate in its state two condition thereby connect-
ing said state two output to said first state two analog
switch, said second state two analog switch, and said third

state two analog switch thereby causing said first state two
variable resistor to be connected in series with said first
state two analog switch and causing both of said devices to be
connected between said output of said first operational ampli-
fier and said negative input of said first operational
amplifier, causing said second state two variable resistor to
be connected in series with said second state two analog switch
and causing both of said devices to be connected between said
output of said second operational amplifier and said negative
input of said second operational amplifier, causing said third
state two variable resistor to be connected in series with
said third state two analog switch and causing both of said
devices to be connected between said gain set inputs of said
instrumentation amplifier, thereby obtaining said correction
signal from said output of said instrumentation amplifier which
causes said process device to operate at a predetermined rapid
speed in the opposite direction.
11. The method defined in Claim 10, wherein the step
of producing a correction signal from said desired value and
said feedback signals while said process controller is in its
third state, includes the steps of causing said state counter
device to operate in its state three condition thereby connect-
ing said state three output to said first state three analog
switch, said second state three analog switch, and said third
state three analog switch thereby causing said first state
three variable resistor to be connected in series with said
first state three analog switch and causing both of said
86

devices to be connected between said output of said first
operational amplifier and said negative input of said first
operational amplifier causing said second state three vari-
able resistor to be connected in series with said second state
three analog switch and causing both of said devices to be
connected between said output of said second operational
amplifier and said negative input of said second operational
amplifier, causing said third state three variable resistor to
be connected in series with said third state three analog
switch and causing both of said devices to be connected be-
tween said gain set inputs of said instrumentation amplifier,
thereby obtaining said correction signal from said output of
said instrumentation amplifier which will look ahead and
attempt to become saturated as soon as a new desired value is
supplied or a process change occurs by utilizing said rate of
change and said error difference, which will remain unchanged
as long as the process being controlled and said desired value
signal both remain unchanged and in a static condition, which
will, if saturated, be brought out of saturation by utilizing
said error difference and said rate of change in a manner to
change said correction signal and value much faster than if
said error difference only were used, and which will, if said
process is in the dynamic condition, be changed in a series of
occurances to a value smaller in magnitude, but of either
polarity, until it arrives at a value related to said condi-
tion it is desired to set said process to.
12. The method defined in Claim 1, wherein the step
of utilizing said correction signal to cause said process to
87

arrive at said desired condition includes the steps of pro-
viding a corrective action circuit, determining the absolute
value of said correction signal, continuously comparing said
absolute value of said correction signal with a deadband
reference value, changing the output of said corrective action
circuit if said absolute value of said correction signal is
above said deadband reference, not changing said output if
said absolute value is between zero and said deadband refer-
ence value, and utilizing said output to cause said process to
arrive at said desired condition.
13. The method defined in Claim 1, wherein the step
of utilizing said correction signal to cause said process to
arrive at said desired condition includes the steps of provid-
ing a corrective action circuit, supplying said correction
signal to an absolute value circuit and to an analog switch
through a scaling device, providing a connection between the
output of said analog switch and an integrator, determining
the absolute value of said correction signal, continuously
comparing the absolute value of said correction signal with a
deadband reference value, causing said analog switch to be
enabled if the absolute value of said correction signal is
above said deadband reference value thereby permitting a
current flow proportional to said correction signal to enter
said integrator and permit a change in the output of said
corrective action circuit causing said analog switch to be
disabled if the absolute value of said correction signal is
between zero and said deadband reference value, thereby
88

permitting no current to flow from said analog switch to said
integrator and permit no change in the output of said cor-
rective action circuit to take place, and utilizing said
output to cause said process to arrive at said desired
condition.
14. The method defined in Claim 1, wherein the step
of utilizing said correction signal to cause said process to
arrive at said desired condition includes the steps of provid-
ing a corrective action circuit, supplying said correction
signal to an absolute value circuit and to a dual analog
switch through a first scaling device, providing connections
between said dual analog switch and a summing integrator,
determining the absolute value of said correction signal,
providing a polarity signal to said dual analog switch through
a second scaling device equal to the polarity of said correc-
tion signal, continuously comparing the absolute value of said
correction signal with a deadband reference value, causing
said dual analog switch to be enabled if the absolute value of
said correction signal is above said deadband reference value
thereby permitting a first current flow proportional to said
correction signal to enter said summing integrator, permitting
a second current flow proportional to said polarity signal to
enter said summing integrator and permitting a change in the
output of said corrective action circuit, causing said dual
analog switch to be disabled if the absolute value of said
correction signal is between zero and said deadband reference
value, thereby permitting no current to flow from said dual
89

D18-1786-54
analog switch to said summing integrator and permitting no change
in the output of said corrective action circuit, and utilizing
said output to cause said process to arrive at said desired
condition.
15. The method defined in Claim 1, wherein the step of
utilizing said correction signal to cause said process to arrive
at said desired condition includes the steps of providing a
corrective action circuit, determining the absolute value of said
correction signal, providing a polarity signal corresponding to
the polarity of said correction signal, continuously comparing
the absolute value of said correction signal with a deadband
reference value, providing a clock signal if the absolute value
of said correction signal is greater than said deadband reference
value, providing no clock output signal if the absolute value of
said correction signal is between zero and said deadband
reference value, and utilizing said clock output and said
polarity signals to cause said process to arrive at said desired
condition.
16. The method defined in Claim 12, 13 or 15, wherein
the step of determining the absolute value of said correction
signal includes the steps of providing a first absolute value
circuit operational amplifier having a positive and negative
input and an output, providing a second absolute value circuit
operational amplifier having a positive and negative input and an
output, connecting said positive input of said first absolute
value circuit operational amplifier to

D18-1786-54
analog common through a resistor having a value of 2/3 R,
connecting said positive input of said second absolute value
circuit operational amplifier to analog common through a resistor
having a value of 2/3 R, connecting said negative input of said
first absolute value circuit operational amplifier to a first
summing junction, supplying said correction signal to a second
summing junction through a resistor having a value of 2R and to
said first summing junction through a resistor having a value of
R, connecting between said first summing junction and said second
summing junction two resistors in series, both having a value of
R, providing a first steering diode having its anode connected to
the junction of said two resistors in series and its cathode
connected to said output of said first absolute value circuit
operational amplifier, providing a second steering diode having
its cathode connected to said first summing junction and its
anode connected to the output of said first absolute value
circuit operational amplifier, connecting the negative input of
said second absolute value circuit operational amplifier to said
second summing junction, connecting the output of said second
absolute value circuit operational amplifier to said second
summing junction through a resistor having a value of 2R, thereby
causing the output of said second absolute value circuit
amplifier to be the absolute value of said correction signal in
which the magnitude is equal to or exceeds zero.
17. The method defined in Claim 15 wherein the steps
of determining the absolute value of said correction signal and
providing a polarity signal corresponding
91

to the polarity of said correction signal includes the steps
of providing a first absolute value circuit operational
amplifier having a positive and negative input and an output,
providing a second absolute value circuit operational amplifier
having a positive and negative input and an output, providing
a third absolute value circuit operational amplifier having a
positive and negative input and an output, connecting said
positive input of said first absolute value circuit opera-
tional amplifier to analog common through a resistor having a
value of 2/3 R, connecting said positive input of said second
absolute value circuit operational amplifier to analog common
through a resistor having a value of 2/3 R, connecting said
negative input of said first absolute value circuit operational
amplifier to a first summing junction, supplying said correc-
tion signal to a second summing junction through a resistor
having a value of 2R and to said first summing junction
through a resistor having a value of R, connecting between
said first summing function and said second summing junction
two resistors in series, both having a value of R, providing a
first steering diode having its anode connected to the junction
of said two resistors in series and its cathode connected to
said output of said first absolute value circuit operational
amplifier, providing a second steering diode having its cathode
connected to said first summing junction and its anode con-
nected to the output of said first absolute value circuit
operational amplifier, connecting the negative input of said
second absolute value circuit operational amplifier to said
second summing junction, connecting the output of said second
92

D18-1786-54
absolute value circuit operational amplifier to said second
summing junction through a resistor having a value of 2R, thereby
causing the output of said second absolute value circuit
amplifier to be the absolute value of said correction signal in
which the magnitude is equal to or exceeds zero, connecting the
output of said first absolute value circuit operational amplifier
to the negative input of said third absolute value circuit
operational amplifier, connecting the postive input of said third
absolute value circuit operational amplifier to analog common
through a resistor having a value of ?, forming a feedback
loop by interposing a resistor having a value of 10R between said
output and said positive input of said third absolute value
circuit operational amplifier and obtaining a polarity signal
from the output of said third absolute value circuit operational
amplifier corresponding to the polarity of said correction
signal.
18. The method defined in Claim 11, 12, or 15, wherein
the step of utilizing said correction signal to cause said
process to arrive at said desired condition includes the steps of
providing a driver, connecting said driver to an operator adapted
to be connected to a process device, and supplying said output
signal to said driver to cause said process to arrive at said
desired condition.
19. The method defined in Claim 15 wherein the step of
utilizing said clock output and said polarity signals to cause
93

said process to arrive at said desired condition includes the
steps of providing a driver, connecting said driver to an
operator adapted to a process device, and supplying said clock
and said polarity signals to said driver to cause said process
to arrive at said desired condition.
20. The method defined in Claim 19, wherein said
operator is in the form of a DC stepping motor, and said driver
is in the form of a stepping motor driver adapted to receive
said clock and said polarity signals to control said operator.
21. The method defined in Claim 20, wherein said
stepping motor driver includes a stepper translator connected to
a quad 5 Amp DC driver and is adapted to receive said clock and
said polarity signals and to control said operator.
22. The method defined in Claim 19, wherein said
operator is in the form of an AC synchronous motor.
23. The method defined in Claim 19, wherein the step of
providing said driver includes providing a two-directional
switched driver.
24. The method defined in Claim 23, wherein the step of
supplying a two-directional switched driver includes the steps
of providing an N assignment device, providing a divide by N
circuit having an input, a preset input, and an output, con-
necting the clock signal to said input, connecting said N
94

assignment device to said preset input, providing a retrig-
gerible timer having an input and an output with said input
connected to said output of said divide by N circuit, providing
a first two input AND gate, providing a second two input AND
gate, connecting the output of said retriggerible timer to one
input each of said first two input and said second two input AND
gates, providing an inverter gate connecting said polarity signal
to the second input of said two input AND gate and to the input
of said invertor gate, connecting the output of said inverter
gate to the second input of said first two input AND gate,
providing a first driver transistor having an emitter, a base
and a collector, connecting said output of said first two input
AND gate to the base of said first driver transistor, providing
a first driver relay having a pair of contact connections,
connecting said collector of said first driver transistor to
said first driver relay, providing a second driver transistor
having an emitter, a base, and a collector, connecting said
output of said second two input AND gate to said base of said
second driver transistor, providing a second driver relay having
an input and a pair of contacts, connecting said collector of
said second driver transistor to said input of said second driver
relay, and connecting the emitter of said first and said second
driver transistors to logic common.
25. The method defined in Claim 2, wherein the step of
providing said feedback signal to said process controller
includes the steps of providing a process measurement device
capable of measuring the current state of the process being

controlled, causing said process measurement device to supply a
process correlate signal related to the current condition of the
process, causing said process correlate signal to either be
directly supplied to said process controller or to a feedback
signal device capable of converting or signal conditioning said
process correlate signal into a signal usable by and directly
supplied to said process controller such that said signal
supplied to said process controller is said feedback signal
related to the current condition of the process being controlled.
26. The method defined in Claim 25, wherein the step of
utilizing said correction signal to cause said process to arrive
at said desired condition includes the steps of providing a
corrective action circuit, determining the absolute value of
said correction signal, continuously comparing said absolute
value of said correction signal with a deadband reference value,
changing the output of said corrective action circuit if said
absolute value of said correction signal is above said deadband
reference, not changing the output of said corrective action
circuit if said absolute value is between zero and said deadband
reference value, and utilizing said output to cause said process
to arrive at said desired condition.
27. The method defined in Claim 25, wherein the step of
utilizing said correction signal to cause said process to arrive
at said desired condition includes the steps of providing a
corrective action circuit, supplying said correction signal to
an absolute value circuit and to an analog switch through a
96

scaling device, providing a connection between the output of
said analog switch and an integrator, determining the abso-
lute value of said correction signal, continuously comparing
the absolute value of said correction signal with a deadband
reference value, causing said analog switch to be enabled if
the absolute value of said correction signal is above said
deadband reference value thereby permitting a current flow
proportional to said correction signal to enter said integra-
tor and permit a change in the output of said corrective
action circuit, causing said analog switch to be disabled if
the absolute value of said correction signal is between zero
and said deadband reference value, thereby permitting no cur-
rent to flow from said analog switch to said integrator and
permit no change in the output of said corrective action
circuit to take place, and utilizing said output to cause
said process to arrive at said desired condition.
28. A three-state four-mode process controller
including means of accepting a desired value signal related
to the condition to which it is desired to set said process,
accepting a feedback signal from the process being controlled
indicating the current condition of the process, accpeting a
reset state signal adapted to insure that the process con-
troller can be easily reset to a state one condition as
desired, which causes said process controller to begin opera-
tion in its first state, means to produce a correction signal
while said process controller is in said first state which
will cause the process device which changes the condition of
97

the process to operate at a predetermined rapid speed in a
first desired direction, utilizing said desired value, feed-
back, and reset state signals to cause said process control-
ler to change to its second state of operation when the error
difference between said desired value and said feedback signal
changes polarity, means to produce a correction signal while
said process controller is in said second state which will
cause said process device to operate at a predetermined rapid
speed in the opposite direction, utilizing said error differ-
ence, the rate of change of said error difference, and the
reset state signal to cause said process controller to change
to its third state of operation when the summation of said
error difference and said rate of change of said error dif-
ference changes polarity, means to produce a correction
signal from said desired value and said feedback signals
while said process controller is in its third state which
will look ahead and attempt to become saturated as soon as a
new desired value is supplied or a process change occurs by
utilizing said rate of change and said error difference,
which will remain unchanged as long as the process being con-
trolled, and said desired value signal both remain unchanged
and in a static condition, which will, if saturated, be
brought out of saturation by utilizing said error difference
and said rate of change in a manner to change said correction
signal in value much faster than if said error difference
only were used, and which will, if said process is in a dyna-
mic condition, be changed in a series of occurrences to a
value smaller in magnitude, but of either polarity, until it
98

arrives at a value related to said condition it is desired to
set said process to, and utilizing said correction signal to
cause said process to arrive at said desired condition.
29. The device defined in Claim 28, wherein said means
to accept said desired value signal and said reset state signal
includes a desired setting device adapted to supply for accept-
ance by said process controller a voltage reference indicating a
desired value and said reset state signal.
30. The device defined in Claim 29, wherein said
desired setting device is a potentiometer and a pushbutton
switch.
31. The device defined in Claim 28, wherein said means
to accept said desired value signal and said reset state signal
includes a desired setting device adapted to supply for accept-
ance by said process controller a reference voltage signal and
said reset state signal and an automation device adapted to
automatically select said reset state signal if desired and
automatically change said reference voltage to a value appropri-
ate to a next condition upon completion of a test.
32. The device defined in Claim 28, wherein said means
to accept a feedback signal includes a process measurement
device adapted to provide a process correlate signal related to
the current condition of the process and to supply for accept-
ance by said process controller a voltage signal indicating a
feedback value related to the current condition of the process.
99

33. The device defined in Claim 28, wherein said means
to accept a feedback signal includes a process measurement
device adapted to provide a process correlate signal related to
the current condition of the process and a feedback signal
device adapted to convert said process correlate signal to a
voltage signal and to supply for acceptance by said process con-
troller said voltage signal indicating a feedback value related
to the current condition of the process.
34. The device defined in Claim 33, wherein said feed-
back signal device is a pressure transducer.
35. The device defined in Claim 28, wherein the means
to produce a correction signal includes a three-state differ-
ential input circuit adapted to determine the error difference
between said desired value signal and said feedback signal,
determine the rate of change of said error difference, and
supply said correction signal related to said reset state
signal and to the algebraic sum of said error difference and
said rate of change of said error difference.
36. The device defined in Claim 35, wherein said
three-state differential input circuit includes a three-state
error and rate amplifier circuit, means to accept said feed-
back signal connected to said three-state error and rate
amplifier circuit, and means to accept said desired value sig-
nal connected to said three-state error and rate amplifier
circuit, means to accept said reset state signal connected to
100

said three-state error and rate amplifier circuit, all adapted
to enable said three-state error and rate amplifier circuit to
provide a correction signal.
37. The device defined in Claim 35, wherein said
three-state differential input circuit includes a three-state
error and rate amplifier circuit, a valid range check circuit,
means to accept said feedback signal connected to said
three-state error and rate amplifier circuit, means to accept
said desired value signal connected to said three-state error
and rate amplifier circuit, means to accept said reset state
signal connected to said three-state error and rate amplifier
circuit, all adapted to enable said three-state error and rate
amplifier circuit to provide a correction signal.
38. The device defined in Claim 35, wherein said
three-state differential input circuit includes a three-state
error and rate amplifier circuit, a valid range check circuit,
a scaling and meter protection circuit, means to accept said
feedback signal connected to said three-state error and rate
amplifier circuit and to said scaling and meter protection
circuits, means to accept said desired value signal connected
to said three-state error and rate amplifier circuit and to
said scaling and meter protection circuit, means to accept
said reset state signal connected to said three-state error
and rate amplifier circuit, all adapted to enable said
three-state error and rate amplifier circuit to provide a cor-
rection signal and to enable said scaler and meter protection
circuit to provide a deviation meter output signal.
101

39. The device defined in Claim 38, wherein said
scaling and meter protection circuit includes a first scaling
operational amplifier having positive and negative inputs and
an output, a second scaling operational amplifier having posi-
tive and negative inputs and an output, said desired value
signal connected to said positive input of said first scaling
operational amplifier, said feedback signal connected to said
positive input of said second scaling operational amplifier, a
first current limiting resistor connected to the output of
said first scaling circuit operational amplifier, a second
current limiting resistor connected to the output of said
second scaling circuit operational amplifier, the negative
input of said first scaling operational amplifier connected to
said first current limiting resistor, the negative input of
said second scaling circuit operational amplifier connected to
said second current limiting resistor, and a scaling resistor
connected to said negative input of said operational amplifier,
all adapted to supply a differential output which is in the
form of voltage and has limited current capacity such that a
meter will not be over ranged.
40. The device defined in any one of Claims 37-38,
wherein said valid range check circuit includes a high limit
comparator having positive and negative inputs and an output,
a low limit comparator having positive and negative inputs and
an output, a high limit set point connected to the positive
input of said high limit comparator, a low limit set point
connected to the positive input of said low limit comparator,
102

D18-1786-54
said desired value signal connectred to the negative input of
said high limit and of said low limit comparator, a high limit
diode having its cathode connected to said output of said high
limit comparator, a low limit diode having its anode connected to
the output of said low limit comparator, the anode of said high
limit diode and the cathode of said low limit diode being
connected together to form the saturation override signal
supplied to said three-state error and rate amplifier circuit,
all adapted to act in a manner to cause the correction signal to
become saturated if said desired value signal is outside said
high limit or said low limit set points, but to operate in a
normal mode supplying said correction signal if said desired
value is within said high limit and said low limit set points.
41. The device defined in Claim 36 wherein said three
state error and rate amplifier circuit includes a first
operational amplifier having positive and negative inputs and an
output, a second operational amplifier having positive and
negative inputs and an output, and instrumentation amplifier
having positive and negative inputs, an output, and gain set
inputs, an edge detector having an input, a pulse output, and a
ground, a state counter device having a reset state input, a
clock input, a clock inhibit input, a state one output, a state
two output, and a state three output, with said reset state
signal to said reset state input of said state counter device,
said polarity signal
103

D18-1786-54
corresponding to the polarity of said correction signal connected
to said input of said edge detector, said saturation override
signal connected to said negative input of said instrumentation
amplifier, said output of said edge detector connected to said
clock input of said state counter device, said ground of said
edge detector being connected to ground, said desired value
signal connected to said positive input of said first operational
amplifier, said feedback signal connected to said positive input
of said second operational amplifier, a capacitor interposed
between said negative inputs of said first and said second
operational amplifiers, said state three output of said state
counter device connected to said clock inhibit input thereof, a
first state three, a second state three and a third state three
analog switch connected to said state three output of said state
counter device, a first state two, a second state two and a third
state two analog switch connected to said state two output of
said state counter device, a first state one, a second state one
and a third state one analog switch connected to said state one
output of said state counter device, said output of said first
operational amplifier connected to said positive input of said
instrumental amplifier, said first state three analog switch,
said first state two analog switch, and said first state one
analog switch connected to the negative input of said first
operational amplifier, a first state three variable resistor
connected between the output of said first operational amplifier
104

D18-1786-54
and said first state three analog switch, a first state two
variable resistor connected between the output of said first
operational amplifier and said first state two analog switch, a
first state one variable resistor connected between the output of
said first operational amplifier and said first state one analog
switch, and said second state three analog switch, said second
state two analog switch, and said second state one analog switch
also connected to the negative input of said second operational
amplifier, a second state three variable resistor connected
between the output of said second operational amplifier and
second state three analog switch, a second state two variable
resistor connected between the output of said second operational
amplifier and said second state two analog switch, a second state
one variable resistor connected between the output of said second
operational amplifier and said second state one analog switch, a
third state three variable resistor and a said third state three
analog switch connected across said gain set inputs of said
instrumentation amplifier in series, a third state two variable
resistor and a third state two analog switch connected across
said gain set inputs of said instrumentation amplifier in series,
a third state one variable resistor and said third state one
analog switch connected across the gain set input of said
instrumenation amplifier in series, a resistor interposed between
said output of said second operational amplifier and said
negative input of said instrumentation amplifier, all adapted to
produce a correction
105

D18-1786-54
signal which will when said state counter device is in its state
one position drives said process device at its rapid determined
speed in a first desired direction, which will, when said state
counter device is in its second state drives said process device
at its rapid predetermined speed in a direction opposite to said
first desired direction, and which will, when said state counter
is in its third state, provide a correction signal as a function
of
<IMG>
wherein:
G = rate plus proportional gain factor
F = feedback signal voltage
DV = desired value signal voltage
C1= capacitance of capacitor C1 in farads
R1= resistance of feedback resistor R1 in ohms
R2= resistance of feedback resistor R2 in ohms
d/dt = derivative of, which respect to time
106

42. The device defined in Claim 41 wherein said edge
detector comprises an exclusive-or gate having two inputs and
an output with the output thereof serving as the output of
said edge detector, a first edge detector resistor interposed
between said input of said edge detector and said first input
of said exclusive-or gate, a second edge detector resistor
interposed between said input of said edge detector and said
second input of said exclusive-or gate, and a capacitor inter-
posed between said second input of said exclusive-or gate and
ground.
43. The device defined in Claim 28, wherein means to
cause the process device to operate includes a corrective
action circuit adapted to provide signals to a driving means
to adjust said process device.
44. The device defined in Claim 43, wherein said cor-
rective action circuit is adapted to provide signals to said
driving means to operate a stepping motor or a reversible
device to adjust said process device and includes an absolute
value circuit having an input adapted to receive said correc-
tion signal and having outputs consisting of a polarity signal
and an absolute value signal equivalent to the absolute value
107

of the correction signal, a deadband comparator having an
input and an output with said input connected to the output of
said absolute value circuit, means to supply deadband refer-
ence values to said deadband comparator, a summing amplifier
having an input and an output, with said input connected to
the absolute value output of said absolute value circuit, a
voltage to frequency converter having an input and an output
with said output in the form of a clock signal,with said input
connected to the output of said summing amplifier, an analog
switch having an input, a control input and an output with
said input being connected to the output of said voltage to
frequency converter and said control input being connected to
the output of said deadband comparator, with said clock signal
being passed through said analog switch and forming a clock
output signal, with said clock output signal and said polarity
signal supplied to said driving means thereby adjusting said
process device.
45. The device defined in Claim 44, wherein said
driving means consists of a stepping motor translator
46. The device defined in Claim 45, wherein said
stepping motor translator consists of a stepper translator
connected to a quad 5ADC driver.
47. The device defined in Claim 44, wherein said
driving means consists of a two directional switched driver.
108

48. The device defined in Claim 43, wherein said
corrective action circuit is adapted to provide signals to a
driving means to operate an operator which is pneumatic in
nature or requires a variable reference signal to adjust said
process device and includes an absolute value circuit having
an input adapted to receive said correction signal and an
output, a scaling circuit having an input and an output, the
input of said scaling circuit also connected to said correction
signal, an analog switch having an input, a control input, and
an output, the input thereof being connected to said output of
said scaling circuit, a deadband comparator having an input, a
reference input, and an output, with said input thereof
connected to said output of said absolute value circuit and
said output of said deadband comparator connected to said
control input of said analog switch, a means to supply
deadband reference values, connected to said reference input
of said deadband comparator, an integrator having an input and
an output with said input thereof being connected to said
output of said analog switch, a buffer-scaler having an input
and an output with said input thereof connected to said output
of said integrator circuit, and said output thereof supplying
a signal to said driving means thereby adjusting said process
device.
49. The device defined in Claim 43, wherein said
corrective action circuit is adapted to provide signals to a
driving means to operate an operator which is pneumatic in
nature or requires a variable reference signal to adjust the
109

D18-1786-54
process device and includes an absolute value circuit having an
input adapted to receive said correction signal, a polarity
output, and an absolute value output a first scaling device
having an input and an output with said input connected to said
corrective signal, with said output connected to said input of
said absolute value circuit, a second scaling device having an
input and an output with the input thereof connected to said
polarity output of said absolute value circuit, a dual analog
switch having its two inputs connected to the outputs of said
first and said second scaling device and having two outputs, a
deadband comparator having an input, a reference input, an
output, with said input thereof connected to said output of said
absolute value circuit, with said output of said deadband
comparator connected to the control input of said dual analog
switch, means to supply said deadband reference values to said
reference input of said deadband comparator, a summing integrator
having two inputs and an output with said inputs thereof
connected to said outputs of said dual analog switch, a
buffer-scaler having an input and an output with said input of
said buffer-scaler connected to the output of said summing
integrator, and the output of said buffer-scaler adapted to
supply a signal to said driving means thereby adjusting said
process device.
50. The device defined in Claim 44 or 48, wherein said
absolute value circuit includes a first absolute value circuit
operational amplifier having positive and negative inputs and an
output, an analog common, a resistor having
110

a value of 2/3 R connected between the positive input of said
first absolute value circuit operational amplifier and analog
common, a first summing junction, a connection between the
negative input of said first absolute value circuit operational
amplifier and said first summing junction, a resistor having a
value of R adapted to receive said correction signal and con-
nected to said first summing junction, a second summing junc-
tion, a resistor of value 2R interposed between said correction
signal and said second summing junction, a junction point, a
first resistor of value R interposed between said junction
point and said first summing junction, a second resistor of
value R interposed between said junction point and said second
summing junction, a first steering diode having an anode and a
cathode with said anode connected to said junction point and
with said cathode connected to the output of said first abso-
lute value circuit operational amplifier, a second steering
diode having an anode and a cathode with its cathode connected
to said first summing junction and its anode connected to the
output of said first operational amplifier, a second absolute
value circuit operational amplifier having positive and nega-
tive inputs and an output, with said negative input connected
to said second summing junction, a resistor having a value of
2/3 R interposed between analog common and the positive input
of said second absolute value operational amplifier, a resistor
of value 2R interposed between said output of said second
absolute value circuit operational amplifier and said second
summing junction, all adapted to provide a signal at the output
of said second absolute value circuit operational amplifier
corresponding to the absolute value of said correction signal.
111

D18-1786-54
51. The device defined in Claim 44, 46 or 47, wherein
said absolute value circuit includes a first absolute value
circuit operational amplifier having positive and negative inputs
and an output, an analog common, a resistor having a value of 2/3
R connected between the positive input of said first absolute
value circuit operational amplifier and analog common, a first
summing junction, a connection between the negative input of said
first absolute value circuit operational amplifier and said first
summing junction, a resistor having a value of R adapted to
receive said correction signal and connected to said first
summing junction, a second summing junction, a resistor of value
2R interposed between said correction signal and said second
summing junction, a junction point, a first resistor of value R
interposed between said junction point and said first summing
junction, a second resistor of value R interposed between said
junction point and said second summing junction, a first steering
diode having an anode and a cathode with said anode connected to
said junction point and with said cathode connected to the output
of said first absolute value circuit operational amplifier, a
second steering diode having an anode and a cathode with its
cathode connected to said first summing junction and its anode
connected to the output of said first operational amplifier, a
second absolute value circuit operational amplifier having
positive and negative inputs and an output, with said negative
input connected to said second summing junction, a resistor
having a value of 2/3 R interposed between analog common and the
positive input of said second absolute value operational
amplifier, a
112

D18-1786-54
resistor of value 2R interposed between said output of said
second absolute value circuit operational amplifier and said
second summing junction, a third absolute value circuit
operational amplifier having positive and negative inputs and an
output, the negative input of said third absolute value circuit
operational amplifier being connected to the output of said first
absolute value circuit operational amplifier, a resistor having a
value of ? connected between the positive input of said third
operational amplifier and analog common, and a resistor of value
10R interposed between the output of said third absolute value
circuit operational amplifier and its positive input, all adapted
to provide a polarity signal at the output of said third absolute
value circuit operational amplifier and to provide a signal
corresponding to the absolute value of said correction signal at
the output of said second absolute value circuit operational
amplifier.
52. The device defined in Claim 44, 46, 47, wherein
said summing amplifier includes an operational amplifier having a
positive and negative input and an output, with said positive
input connected to analog common through a resistor, a voltage
follower circuit including the resistor of value Rf interposed
between said output and negative input of said operational
amplifier, an adjustable resistor of value Rb connected to
the negative input of said operational amplifier and adapted to
receive said absolute value signal, and a base speed reference
113

D18-1786-54
device also being connected through a resistor of value Ra to
said negative input of said operational amplifier, all adapted
to produce an output from said operational amplifier according to
the function <IMG>
wherein:
Rb = summing amplifier maximim speed input resistor in ohms
Ra = summing amplifier base speed reference resistor in ohms
Y = base speed reference voltage
X = correction signal
Rf = summing amplifier feedback resistor in ohms
114

53. The device defined in Claim 48, wherein said
integrator includes an integrator operational amplifier having
positive and negative inputs and an output with said positive
input of said integrator operational amplifier being connected
to analog common, a capacitor being connected from said negative
input of said integrator operational amplifier to said output
thereof, and a resistor being connected to said negative input
of said operational amplifier, alladapted to provide an output
to said buffer-scaler.
54. The device defined in Claim 49, wherein said
summing integrator includes a summing integrator operational
amplifier having positive and negative inputs and an output with
the positive input of said summing integrator operational ampli-
fier being connected to analog common, a pair of resistors con-
nected to the negative input of said summing integrator opera-
tional amplifier, and a capacitor interposed between said
negative input of said summing integrator operational amplifier
and said output, all adapted to provide an output to said
buffer-scaler.
55. The device defined in any one of claims 48-49,
wherein said buffer-scaler includes an NPN transistor, a PNP
transistor, the output signal from the integrator or summing
115

integrator supplied to the base of both transistors, the col-
lector of said NPN transistor connected to the positive power
supply voltage, the collector of said PNP transistor connected
to negative power supply voltage, and the emitters of both
transistors connected to a scaling resistance Rs which provides
an output signal to said driving means.
56. The device defined in Claim 47, wherein the two-
directional switched driver consists of a divide by N circuit
having an input, a preset input, and an output, said clock
output signal from said corrective action circuit being con-
nected to said input of said divide by N circuit, an N assign-
ment device connected to said preset input of said divide by N
circuit, a retriggerible timer having an input and an output
with said output of said divide by N circuit being connected to
said input of said timer, a first two input AND gate and a
second two input AND gate, said output of said timer connected
to one input each of said first and said second two input AND
gates, said polarity signal from said corrective action circuit
connected to said second input of said second two input AND
gate, an inverter having an input and an output with said
polarity signal from said corrective action circuit also being
connected to said input of said inverter, and said output of
said inverter being connected to said second input of said first
two input AND gate, a first driver transistor having an emitter,
a base and a collector, said output of said first two input AND
gate connected to said base of said first driver transistor,
said emitter of said first driver transistor connected to logic
116

common, said output of said second two input AND gate con-
nected to the base of said second driver transistor, said
emitter of said second driver transistor being connected to
logic common, a first driver relay connected to said col-
lector of said first driver transistor, a second driver relay
connected to said collector of said second driver transistor,
and contact connections provided on said first driver relay
and on said second driver relay for operating said operator
thereby adjusting said process device.
57. The device defined in Claim 29, wherein said
means to accept said feedback signal includes a process
measurement device adapted to provide a process correlate
signal related to the current condition of the process and a
feedback signal device adapted to convert said process cor-
relate signal to a voltage signal and to supply for accept-
ance by said process controller said voltage signal indicat-
ing a feedback value related to the current condition of the
process.
58. The device defined in Claim 29, wherein the
means to produce a correction signal includes a three-state
differential input circuit adapted to determine the error
difference between said desired value signal and said feed-
back signal, determine the rate of change of said error
difference, and supply said correction signal related to said
reset state signal and to the algebraic sum of said error
difference and said rate of change of said error difference.
117

59. The device defined in Claim 58, wherein said
three-state differential input circuit includes an three-
state error and rate amplifier circuit, a valid range check
circuit, a scaling and meter protection circuit, means to
accept said feedback signal connected to said three-state
error and rate amplifier circuit and to said scaling and
meter protection circuits, means to accept said desired value
signal connected to said three-state error and rate amplifier
circuit and to said scaling and meter protection circuit,
means to accept said reset state signal connected to said
three-state error and rate amplifier circuit, all adapted to
enable said three-state error and rate amplifier circuit to
provide a correction signal and to enable said scaler and
meter protection circuit to provide a deviation meter output
signal.
60. The device defined in Claim 59, wherein said
scaling and meter protection circuit includes a first scaling
operational amplifier having positive and negative inputs and
an output, a second scaling operational amplifier having pos-
itive and negative inputs and an output, said desired value
signal connected to said positive input of said first scaling
operational amplifier, said feedback signal connected to said
positive input of said second scaling operational amplifier,
a first current limiting resistor connected to the output of
said first scaling circuit operational amplifier, a second
current limiting resistor connected to the output of said
second scaling circuit operational amplifier, the negative
118

input of said first scaling operational amplifier connected
to said first current limiting resistor, the negative input
of said second scaling circuit operational amplifier con-
nected to said second current limiting resistor, and a
scaling resistor connected to said negative input of said
operational amplifier, all adapted to supply a differential
output which is in the form of voltage and has limited
current capacity such that a meter will not be over ranged.
61. The device defined in Claim 60, wherein said
valid range check circuit includes a high limit comparator
having positive and negative inputs and an output, a low
limit comparator having positive and negative inputs and an
output, a high limit set point connected to the positive
input of said high limit comparator, a low limit set point
connected to the positive input of said low limit comparator,
said desired value signal connected to the negative input of
said high limit and of said low limit comparator, a high
limit diode having its cathode connected to said output of
said high limit comparator, a low limit diode having its
anode connected to the output of said low limit comparator,
the anode of said high limit diode and the cathode of said
low limit diode being connected together to form the satura-
tion override signal supplied to said three-state error and
rate amplifier circuit, all adapted to act in a manner to
cause the correction signal to become saturated if said
desired value signal is outside said high limit or said low
limit set points, but to operate in a normal mode supplying
119

said correction signal if said desired value is within said
high limit and said low limit set points.
62. The device defined in Claim 61, wherein said
three state error and rate amplifier circuit includes a first
operational amplifier having positive and negative inputs and
an output, a second operational amplifier having positive and
negative inputs and an output, an instrumentation amplifier
having positive and negative inputs, an output, and gain set
inputs, an edge detector having an input, a pulse output, and
a ground, a state counter device having a reset state input,
a clock input, a clock inhibit input, a state one output, a
state two output, and a state three output, with said reset
state signal to said reset state input of said state counter
device, said polarity signal corresponding to the polarity of
said correction signal connected to said input of said edge
detector, said saturation override signal connected to said
negative input of said instrumentation amplifier, said output
of said edge detector connected to said clock input of said
state counter device, said ground of said edge detector being
connected to ground, said desired value signal connected to
said positive input of said first operational amplifier, said
feedback signal connected to said positive input of said
second operational amplifier, a capacitor interposed between
said negative inputs of said first and said second opera-
tional amplifiers, said state three output of said state
counter device connected to said clock inhibit input thereof,
a first state three, a second state three and a third state
120

three analog switch connected to said state three output of
said state counter device, a first state two, a second state
two and a third state two analog switch connected to said
state two output of said state counter device, a first state
one, a second state one and a third state one analog switch
connected to said state one output of said state counter
device, said output of said first operational amplifier con-
nected to said positive input of said instrumentation
amplifier, said first state three analog switch, said first
state two analog switch, and said first state one analog
switch connected to the negative input of said first opera-
tional amplifier, a first state three variable resistor
connected between the output of said first operational
amplifier and said first state three analog switch, a first
state two variable resistor connected between the output of
said first operational amplifier and said first state two
analog switch, a first state one variable resistor connected
between the output of said first operational amplifier and
said first state one analog switch, said second state three
analog switch, said second state two analog switch, and said
second state one analog switch also connected to the negative
input of said second operational amplifier, a second state
three variable resistor connected between the output of said
second operational amplifier and second state three analog
switch, a second state two variable resistor connected
between the output of said second operational amplifier and
said second state two analog switch, a second state one var-
iable resistor connected between the output of said second
121

operational amplifier and said second state one analog
switch, a third state three variable resistor and said third
state three analog switch connected across said gain set
inputs of said instrumentation amplifier in series, a third
state two variable resistor and a third state two analog
switch connected across said gain set inputs of said instru-
mentation amplifier in series, a third state one variable
resistor and said third state one analog switch connected
across the gain set input of said instrumentation amplifier
in series, a resistor interposed between said output of said
second operational amplifier and said negative input of said
instrumentation amplifier, all adapted to produce a correc-
tion signal which will when said state counter device is in
its state one position drives said process device at its
rapid determined speed in a first desired direction, which
will, when said state counter device is in its second state
drives said process device at its rapid predetermined speed
in a direction opposite to said first desire direction, and,
which will, when said state counter is in its third state,
122

D2-1786-54
62 contd.
provide a correction signal as a function of
<IMG>
wherein:
G = R1 x C1
F = feedback signal voltage
DV = desired value signal voltage
C1= capacitance of capacitor C1 in farads
R1= resistance of feedback resistor R1 in ohms
R2= resistance of feedback resistor R2 in ohms
d/dt= derivative of, with respect to time
63. The device defined in Claim 62, wherein means to
cause the process device to operate includes a corrective action
circuit adapted to provide signals to a drive means to adjust
said process device.
123

64. The device defined in Claim 63, wherein said
corrective action circuit is adapted to provide signals to
said driving means to operate a stepping motor or a reversible
device to adjust said process device and includes an absolute
value circuit having an input adapted to receive said correc-
tion signal and having outputs consisting of a polarity signal
and an absolute value signal equivalent to the absolute value
of the correction signal, a deadband comparator having an
input and an output with said input connected to the output of
said absolute value circuit, means to supply deadband refer-
ence values to said deadband comparator, a summing amplifier
having an input and an output, with said input connected to
the absolute value output of said absolute value circuit, a
voltage to frequency converter having an input and an output
with said output in the form of a clock signal,with said input
connected to the output of said summing amplifier, an analog
switch having an input, a control input and an output with
said input being connected to the output of said voltage to
frequency converter and said control input being connected to
the output of said deadband comparator, with said clock signal
being passed through said analog switch and forming a clock
output signal, with said clock output signal and said polarity
signal supplied to said driving means thereby adjusting said
process device.
65. The device defined in Claim 64, wherein said
driving means consists of a a stepping motor translator.
124

66. The device defined in Claim 63, wherein said
corrective action circuit is adapted to provide signals to a
driving means to operate an operator which is pneumatic in
nature or requires a variable reference signal to adjust the
process device and includes an absolute value circuit having
an input adapted to receive said correction signal, a polarity
output, and an absolute value output a first scaling device
having an input and an output with said input connected to
said corrective signal, with said output connected to said
input of said absolute value circuit, a second scaling device
having an input and an output with the input thereof connected
to said polarity output of said absolute value circuit, a
dual analog switch having its two inputs connected to the
outputs of said first and said second scaling device and
having two outputs, a deadband comparator having an input, a
reference input, an output, with said input thereof connected
to said output of said absolute value circuit, with said
output of said deadband comparator connected to the control
input of said dual analog switch, means to supply said
deadband reference values to said reference input of said
deadband comparator, a summing integrator having two inputs
and an output with said inputs thereof connected to said
outputs of said dual analog switch, a buffer-scaler having an
input and an output with said input of said buffer-scaler
connected to the output of said summing integrator, and the
output of said buffer-scaler adapted to supply a signal to
said driving means thereby adjusting said process device.
125

67. The device defined in any one of Claims 64-66,
wherein said absolute value circuit includes a first absolute
value circuit operational amplifier having positive and
negative inputs and an output, an analog common, a resistor
having a value of 2/3 R connected between the positive input
of said first absolute value circuit operational amplifier and
analog common, a first summing junction, a connection between
the negative input of said first absolute value circuit opera-
tional amplifier and said first summing junction, a resistor
having a value of R adapted to receive said correction signal
and connected to said first summing junction, a second summing
junction, a resistor of value 2R interposed between said cor-
rection signal and said second summing junction, a junction
point, a first resistor of value R interposed between said
junction point and said first summing junction, a second re-
sistor of value R interposed between said junction point and
said second summing junction, a first steering diode having an
anode and a cathode with said anode connected to said junction
point and with said cathode connected to the output of said
first absolute value circuit operational amplifier, a second
steering diode having an anode and a cathode with its cathode
connected to said first summing junction and its anode con-
nected to the output of said first operational amplifier, a
second absolute value circuit operational amplifier having
positive and negative inputs and an output, with said negative
input connected to said second summing junction, a resistor
having a value of 2/3 R interposed between analog common and
the positive input of said second absolute value operational
126

amplifier, a resistor of value 2R interposed between said out-
put of said second absolute value circuit operational ampli-
fier and said second summing junction, a third absolute value
circuit operational amplifier having positive and negative
inputs and an output, the negative input of said third
absolute value circuit operational amplifier being connected
to the output of said first absolute value circuit operational
amplifier, a resistor having a value of ? connected between
the positive input of said third operational amplifier and analog
common, and a resistor of value 10R interposed between the output
of said third absolute value circuit operational amplifier and
its positive input, all adapted to provide a polarity signal at
the- output of said third absolute value circuit operational
amplifier and to provide a signal corresponding to the absolute
value of said correction signal at the output of said second
absolute value circuit operational amplifier.
68. The device defined in Claim 64, wherein said
summing amplifier includes an operational amplifier having a
positive and negative input and an output, with said positive
input connected to analog common through a resistor, a voltage
follower circuit including the resistor of value Rf interposed
between said output and said negative input of said operational
amplifier, an adjustable resistor of value Rb connected to the
negative input of said operational amplifier and adapted So
receive said absolute value signal, and a base speed reference
device also being connected through a resistor of value Ra to
127

D2-1786-54
68 contd.
said negative input of said operational amplifier, all adapted to
produce an output from said operational amplifier according to
the function -Rf Y + IXI
Ra Rb
wherein:
Rb = summing amplifier maximum speed input resistor in ohms
Ra = summing amplifier base speed reference resistor in ohms
Y = base speed reference voltage
X = correction signal
G = rate plus proportional gain factor
Rf = summing amplifier feedback resistor in ohms
128

69. The device defined in Claim 66, wherein said sum-
ming integrator includes a summing integrator operational ampli-
fier having positive and negative inputs and an output with the
positive input of said summing integrator operational amplifier
being connected to analog common, a pair of resistors connected
to the negative input of said summing integrator operational
amplifier, and a capacitor interposed between said negative
input of said summing integrator operational amplifier and said
output, all adapted to provide an output to said buffer-scaler.
70. The device defined in Claim 69, wherein said buffer-.
scaler includes an NPN transistor, a PNP transistor, the output
signal from the integrator or summing integrator supplied to the
base of both transistors, the collector of said NPN transistor
connected to the positive power supply voltage, the collector of
said PNP transistor connected to negative power supply voltage,
and the emitters of both transistors connected to a scaling
resistance Rs which provides an output signal to said driving
means
71. A method of testing carburetors at any desired num-
ber of points in the carburetors operating range using subsonic
flow to determine the air flow and fuel flow rate through the
test carburetor, said method including the steps providing a
129

D18-1786-54
suitable testing stand on which to mount the carburetor,
providing a suitable hood above said testing stand adapted to
sealingly enclose said carburetor, continuously controlling the
pressure within said hood utilizing a hood pressure measurement
and control system including a three-state four-mode process
controller utilizing desired value, feedback, and reset state
signals to cause said process controller to operate in said three
states depending on the value of said signals so as to quickly
produce the desired hood pressure at each point at which said
carburetor test will take place in the best possible time,
simultaneously controlling the pressure of the fuel entering the
carburetor, simultaneously inducing air flow through said
carburetor by providing a vaccum downstream of said carburetor,
simultaneously determining the flow rate of air and fuel entering
the carburetor, and simultaneously controlling the rotation of
the carburetor throttle plate until the desired predetermined
test condition is achieved.
72. A method of testing carburetors at any desired
number of points in the carburetors operating range using
subsonic flow to determine the air flow and fuel flow rate
through the test carburetor, said method including the steps
providing a suitable testing stand on which to mount the
carburetor, providing a suitable hood above said testing stand
adapted to sealingly enclose said carburetor, continuously
controlling the pressure within said hood, simultaneously
130

D18-1786-54
controlling the pressure of the fuel entering the carburetor,
simultaneously inducing air flow through said carburetor by
continuously controlling the manifold vaccum across the
carburetor utilizing a manifold vacuum measurement and control
system including a three-state four-mode process controller,
utilizing desired value, feedback, and reset state signals to
cause said process controller to operate in said three states
depending on the value of said signals, simultaneously
determining the flow rate of air and fuel entering the
carburetor, and simultaneously controlling the rotation of the
carburetor throttle plate until the desired predetermined test
condition is achieved.
73. A method of testing carburetors at any desired
number of points in the carburetors operating range using
subsonic flow to determine the air flow and fuel flow rate
through the test carburetor, said method including the steps
providing a suitable testing stand on which to mount the
carburetor, providing a suitable hood above said testing stand
adapted to sealingly enclose said carburetor, continuously
controlling the pressure within said hood, simultaneously
controlling the pressure of the fuel entering the carburetor,
simultaneously inducing air flow through said carburetor by
providing a vaccum downstream of said carburetor, simultaneously
determining the flow rate of air and fuel entering the
carburetor, and simultaneously controlling the rotation of the
131

D18-1786-54
carburetor throttle plate by the use of an air flow measurement
and control system including a three-state four-mode process
controller until the desired predetermined test condition is
achieved.
74. A method of testing carburetors at any desired
number of points in the carburetors operating range using
subsonic flow to determine the air flow and fuel flow rate
through the test carburetor, said method including the steps
providing a suitable testing stand on which to mount the
carburetor, providing a suitable hood above said testing stand
adapted to sealingly enclose said carburetor, continuously
con-trolling the pressure within said hood, simultaneously
controlling the pressure of the fuel entering the carburetor
by utilizing a three-state four-mode process controller,
simultaneously inducing air flow through said carburetor by
providing a vaccum downstream of said carburetor, simultaneously
determining the flow rate of air and fuel entering the
carburetor, and simultaneously controlling the rotation of the
carburetor throttle plate until the desired predetermined test
condition is achieved.
75. A method or testing carburetors at any desired
numer of points in the carburetors operating range using subsonic
flow to determine the air flow and fuel flow rate through the
test carburetor, said method including the steps providing a
suitable testing stand on which to mount the carburetor, pro-
132

D18-1786-54
viding a suitable hood above said testing stand adapted to
sealingly enclose said carburetor, continuously controlling the
pressure within said hood utilizing a hood pressure measurement
and control system including a three-state four-mode process
controller so as to quickly produce the desired hood pressure at
each point at which said carburetor test will take place in the
best possible time, simultaneously controlling the pressure of
the fuel entering the carburetor, simultaneously inducing air
flow through said carburetor by continuously controlling the
manifold vaccum across the carburetor utilizing a manifold vaccum
measurement and control system including a three-state four-mode
process controller, simultaneously determining the flow rate of
air and fuel entering the carburetor, and simultaneously
controlling the rotation of the carburetor throttle plate until
the desired predetermined test condition is achieved.
76. A method of testing carburetors at any desired
number of points in the carburetors operating range using
subsonic flow to determine the air flow and fuel flow rate
through the test carburetor, said method including the steps of
providing a suitable testing stand on which to mount the
carburetor, providing a suitable hood above said testing stand
adapted to sealingly enclose said carburetor, continuously
controlling the pressure within said hood utilizing a hood
pressure measurement and control system including a three-state
foue-mode process controller so as to quickly produce the desired
133

D18-1786-54
hood pressure at each point at which said carburetor test will
take place in the best possible time, simultaneously controlling
the pressure of the fuel entering the carburetor, simultaneously
inducing air flow through said carburetor by continuously
controlling the manifold vaccum across the carburetor utilizing a
manifold vacuum measurement and control system including a
three-state four-mode process controller, simultaneously
determining the flow rate of air and fuel entering the
carburetor, and simultaneously controlling the rotation of the
carburetor throttle plate by the use of an air flow measurement
and control system including a three-state four-mode process
controller until the desired predetermined test condition is
achieved.
77. The method defined in Claim 75, and including the
step of determining the mass air flow rate entering the
carburetor .
134

78. The method defined in Claim 77, and including the
step of determining the mass fuel flow rate entering the
carburetor.
79. The method defined in Claim 78, and including the
step of calculating the air/fuel ratio from the values of mas
air flow and mass fuel flow previously determined.
80. The method defined in Claim 79, with the carburetor
system being used in a controlled environment room and keeping
the pressure of the air entering said laminar flow tubes
constant.
81. The method defined in Claim 79, with the carburetor
test system drawing air from an air supply system having con-
trolled temperature, pressure and humidity and keeping the pres-
sure of the air entering the system constant.
82. The method defined in Claim 79, wherein the deter-
mining of the mass fuel flow rate includes the steps of providing
a fuel supply, passing the fuel through a mass fuel flow trans-
ducer enroute to the carburetor, measuring the differential pres-
sure across the fuel flow transducer and calculating the actual
mass fuel flow rate from the differential pressure.
83. The method defined in Claim 82, wherein said mass
fuel flow transducer and differential pressure transducer is
replaced by a volumetric flow transducer and including the steps
135

of measuring the temperature of the fuel flowing to said
carburetor and calculating the mass fuel flow rate from said
measured values.
84. The method defined in Claim 82, wherein the mass
fuel flow transducer is replaced by a set of orifices, and the
differential pressure is measured by a differential pressure
transducer and including the steps of measuring the temperature
of the fuel entering the carburetor and calculating the mass
fuel flow rate from said measured values.
85. The method defined in Claim 86, wherein the measur-
ing of the actual fuel pressure entering the carburetor is
performed by measuring the differential pressure between said
transducer and the air pressure inside said test chamber and
calculating the fuel pressure from said measurements.
86. The method defined in Claim 82, and including the
steps of measuring and calculating manifold vacuum across said
carburetor.
87. The method defined in Claim 86, and including the
step of providing a conduit having an inlet and an outlet, con-
necting the inlet of said conduit to said hood and connecting
the outlet of said conduit to said hood pressure measurement and
control system.
136

D18-1786-54
88. The method defined in Claim 87 wherein a process
speed improvement device is included in said hood pressure
measurement and control system.
89. The method defined in Claim 88, wherein the step
of providing said process speed improvement device includes the
step of providing an inline valve and connecting a valve
operator to said inline valve adapted to be controlled by
said hood pressure measurement and control system.
90. An apparatus for testing carburetors at any
desired number of points in the carburetors operating range using
subsonic flow to determine the air flow and fuel flow rate
through the test carburetor, said apparatus including means to
provide a suitable testing stand on which to mount the
carburetor, means to provide a suitable hood above said testing
stand adapted to sealingly enclose said carburetor, means to
continuously control the pressure within said hood utilizing a
hood pressure measurement and control system including a
three-state four-mode process controller so as to quickly produce
the desired hood pressure at each point at which said carburetor
test will take place at the best possible time, means to
simultaneously control the pressure of the fuel entering the
carburetor, means to simultaneously induce air flow through said
carburetor, means to simultaneously determine the flow rate of
air and fuel entering the carburetor, and means to simultaneously
control the rotation of the carburetor throttle plate until the
desired predetermined test condition is achieved.
137

D18-1786-54
91. An apparatus for testing carburetors at any
desired number of points in the carburetors operating range using
subsonic flow to determine the air flow and fuel flow rate
through the test carburetor, said apparatus including means to
provide a suitable testing stand on which to mount the
carburetor, means to provide a suitable hood above said testing
stand adapted to sealingly enclose said carburetor, means to
continuously control the pressure within said hood, means to
simultaneously control the pressure of the fuel entering the
carburetor, means to simultaneously induce air flow through said
carburetor by continuously controlling the manifold vacuum across
the carburetor utilizing a manifold vacuum measurement and
control system including a three-state four-mode process
controller, means to simultaneously determine the flow rate of
air and fuel entering the carburetor, and means to simultaneously
control the rotation of the carburetor throttle plate until the
desired predetermined test condition is achieved.
92. An apparatus for testing carburetors at any desired
number of points in the carburetors operating range using
subsonic flow to determine the air flow and fuel flow rate
through the test carburetor, said apparatus including means to
provide a suitable testing stand on which to mount the
carburetor, means to provide a suitable hood above said testing
stand adapted to sealingly enclose said carburetor, means to
continuously control the pressure within said hood, means to
simultaneously control the pressure of the fuel entering the
carburetor, means to simultaneously induce air flow through
138

D18-1786-54
said carburetor, means to simultaneously determine the flow rate
of air and fuel entering the carburetor, and means to
simultaneously control the rotation of the carburetor throttle
plate by the use of an air flow measurement and control system
including a three-state four-mode process controller until the
desired predetermined test condition is achieved.
93. An apparatus for testing carburetors at any
desired number of points in the carburetors operating range using
subsonic flow to determine the air flow and fuel flow rate
through the test carburetor, said apparatus including means to
provide a suitable testing stand on which to mount the
carburetor, means to provide a suitable hood above said testing
stand adapted to sealingly enclose said carburetor, means to
continuously control the pressure within said hood, means to
simultaneously control the pressure of the fuel entering the
carburetor by utilizing a three-state four-mode process
controller, means to simultaneously induce air flow through said
carburetor, means to simultaneously determine the flow rate of
air and fuel entering the carburetor, and means to simultaneously
control the rotation of the carburetor throttle plate until the
desired predetermined test condition is achieved.
94. An apparatus for testing carburetors at any
desired number of points in the carburetors operating range using
subsonic flow to determine the air flow and fuel flow rate
through the test carburetor, said apparatus including means to
139

D18-1786-54
provide a suitable testing stand on which to mount the
carburetor, means to provide a suitable hood above said testing
stand adapted to sealingly enclose said carburetor, means to
continuously control the pressure within said hood utilizing a
hood pressure measurement and control system including a
three-state four-mode process controller so as to quickly produce
the desired hood pressure at each point at which said carburetor
test will take place in the best possible time, means to
simultaneously control the pressure of the fuel entering the
carburetor, means to simultaneously induce air flow through said
carburetor by continuously controlling the manifold vacuum across
the carburetor utilizing a manifold vacuum measurement and
control system including a three-state four-mode process
controller, means to simultaneously determine the flow rate of
air and fuel entering the carburetor, and means to simultaneously
control the rotation of the carburetor throttle plate until the
desired predetermined test condition is achieved.
95. An apparatus for testing carburetors at any
desired number of points in the carburetors operating range using
subsonic flow to determine the air flow and fuel flow rate
through the test carburetor, said apparatus including means to
provide a suitable testing stand on which to mount the
carburetor, means to provide a suitable hood above said testing
stand adapted to sealingly enclose said carburetor, means to
continuously control the pressure within said hood utilizing a
140

D18-1786-54
hood pressure measurement and control system including a
three-state four-mode process controller so as to quickly produce
the desired hood pressure at each point at which said carburetor
test will take place in the best possible time, means to
simultaneously control the pressure of the fuel entering the
carburetor, means to simultaneously induce air flow through said
carburetor by continuously controlling the manifold vacuum across
the carburetor utilizing a manifold vacuum measurement and
control system including a three-state four-mode process
controller, means to simultaneously determine the flow rate of
air and fuel entering the carburetor, and means to simultaneously
control the rotation of the carburetor throttle plate by the use
of an air flow measurement and control system including a
three-state four-mode process controller until the desired
predetermined test condition is achieved.
96. The apparatus defined in Claim 94, and including
means to determine the mass fuel flow rate entering the
carburetor.
97. The apparatus defined in Claim 96, and including
means to determine the mass air flow rate entering the
carburetor.
98. The apparatus defined in Claim 97, and including
means to calculate the air/fuel ratio of said carburetor from the
values of mass air flow and mass fuel flow.
99. The apparatus as defined in Claim 98, wherein the
141

D18-1786-54
means to induce an air flow through the inlet of said chamber
including a vacuum producing means, a first conduit connected to
an air supply controlled as to temperature, pressure and
humidity, an enlarged chamber having an inlet and an outlet, with
the inlet thereof connected to said first conduit, a second
conduit connected to said outlet with the other end of said
second conduit communicating with said test chamber, a wall
dividing said enlarged chamber into two portions and at least one
flow restricting device mounted through said wall to allow air to
pass through said chamber, an air flow differential pressure
transducer to sense the pressure drop across said flow
restricting device and to provide a signal related to said
pressure drop, means to obtain the absolute pressure upstream of
said flow restricting device, means to sense the temperature
upstream of said flow restricting device, means to calculate from
the differential pressure, absolute pressure and temperature the
actual mass flow rate of air passing through said flow
restricting device.
100. The apparatus as defined in any one of Claims 90
or 94-95, and including a conduit having an inlet and an outlet,
said inlet of said conduit connected to said sealed space under
said hood, said outlet of said conduit being connected to said
hood pressure measurement and control system.
101. The apparatus defined in Claim 94, 95, wherein a
process speed improvement device is included in said hood
pressure measurement and control system.
142

102. The apparatus defined in Claim 101, and including
an operator connected to said process speed improvement device,
a driver connected to said operator and said hood pressure
measurement and control system.
103. The apparatus defined in Claim 102, wherein said
process speed improvement device is in the form of a valve.
104. The device defined in Claim 103, wherein said valve
is adapted to snap shut as soon as said hood pressure measure-
ment and control system is caused to go to its first state.
143

Description

m
r
'r~Je present ~pplication rela~eg ~o process controll~rs ~nd more
partlcularly to an improve~ prQc~as controller whereln the controlllng of
productlon type proc~es ~ more accuratc and ~aster th~n w1th those
controllers pre~e1l~ly ava1lab1e.
W~ have lon~g ~ell lnvolved ln th~ proce~ collcroller art by vlr-
tu~ of the need to quickly and accurately control proc~sse~ involv~d in
stands ~or the te~t1ng of carburctors, ~uch a~ chose discloY~d ln the U.S.
Patent NoY. 3, S17, 552; 3, 524, 344; 3, 8~1, 523; 3, 896D 670; 3, 975, 9S3 an~
4~030,351. Proce6aes whlcll mu~t be controlled ~n the c~rburetor tes11ng
stnnds dlsclosed ln the above pacents are hood pres~ur~, m~nlfold vacuum,
. . i nnd fuel pre~sure, among ot11~rs. When controllirl~ manlfold vacuum, t1~e
concrol of the throctle plate of ~he carburetor ~o ~rlng lt to a de~ir~d posl-
clon to produce a desired manifold vacuurn 1B mo~t crlclcal. In che early
~iays OI carbure~or testing wher~ perhap~ one or two test point~ w~xe
lnvolved, ~ncl ~ccuracy requlr~n~ent~ were low, test t1rne was not a par-
tlcularly lmportan~ fnctor,, ~lowever, v.ith ~he present day en~p11EIsls on
fuel econo.~ly and exh~ust emis~ions, and the need to te8t automoblle c~r ~
buretors .lt many po11lts withln the~r oper~lonal rangc, the ~1llty to nloYe
the c~arbureto~ throt~l~ plate, an~ thus produce a d~lre~ rnanifold Yacuum
~t many test po~nt~ quickly and accura~ely, 18 becomin~ incre~in~ly
important.
- During t1le t~me ~1~el~ accurncy requlr~rnents p~rm1tte~ a ~impl~
set Q~ relay cont~cts operacirlg a motor to onuse the throttle plate ~o move
~rom one poQ1tlon, ~uch ~ off~ldle, to ano~her positlon, ~uch a part
throttle, cornplex controls were not needed. However, ~s te9tY became
-2~
~k;
... . . . .

~fl~'7~7
more comp11cated and accur~cy requirements became tighter, a search
wa~ made to determine a better way to cau~e the movement of the throttle
plate from one poqltlon ~o another.
The idea of using a motor whlch could be moved in gross amounts
clockwi~e and counterclockwise, such as by relay contacts, was aban-
doned, and che use of a motor v~hich could be moved at cwo dlfferent speeds
and could be 8hut off once the proces~ was at or close to its desired value~
called dead band, was lnstituted. Thus, the motor would mo,re at a fast
r~te of speed when the process was far away from the desired valuet and
move at a much ~lower race of speed when the process was near the deslred
value. However, as much of an ~dvance as this t~o-speed throttle drive or
process controller actually was over the prior art, it too was soon too 910w
for the ever increasing demands of production processeQ. This v.~as prl-
marlly because thele were only two fixed speeds, and if ~he process under-
went rapid change, there would be qui~e a tilne lag for the throttle conroller
to ad~ust the throttle plate to a new condltioll within the dead band lirnits,
whlch were l~eco~ng ~maller becau~e of still tighter accuracy requirements.
Therefore~ further experimentation led to the inventloll of a throttle
drive for a carburetor test stand haYing a proportlonal speed feature, in
uhich the ~peed of the driving motor was proportional to the amount of error
in the proces~. Th~s invention, of whlch one of the co-inventors in the
present ca~e was a co-inventor, led to the grant of the U. S. Patent
No. 3,975,953" and it wa~ thought that a~ long last one of the major prob-
lem~ ln the carburetor industry was solved.
-3

~ 6~777
Between the tlme of making that hlvention, and tlle present day,
lt was found that In laboratory carburetor test benches where actual
value6 for production tests of carburetors are detersnined, lt uas deslr-
al:le to improve tlle speed and accuracy of the tests where, ln addition to
throttle control, manifold vacuum and carburetor lnlet pressure control
(known as hood pressure) are also required, At that tirne, such control
of manifold vacuum and hood pressure was done using conventional
proces6 controllers, while tllroctle control uas normally performed man-
ually by the test stand operator. It was found that with the use of a corn-
puter lt was possible to effectively use process control utllizin~ optimum
rate, reset and proportlonal values for all three parameters -- throttle,
manifold vacuum and hood pressure, and becau~e of the ded~catlon of the
computer to one stand, not only would you get the laboratory type accuracy
which was desired, but also the testing speed becnme faster. This lnven-
tion led to the grant of U.S. Patent No. 4,030,351 for Method and Apparatus
for Production Testing of Carburetors by one of the co-inventorsO
During the years that u~ere passinL~ l~y ~ ile these de-relopments
were taking place, the demand for even faster and more accurate produc-
tion test stands were being made, a~d we were con pelled to en~bark on
fur~her research ~o see if we could no~ get a ~ime for ~ typical carburetor
test belov~ the current test time for a particular model carbure~or of
approximately 9 minutes~ and at the same time get the accuracy glven by
our laboratory te6t stands previously mentloned.
The mere implementation of the sIIe~hod used in our la~oratory
te~t stands mi~ht sufflce to solve this serious problesn in the art, How-
,, , . , , , . ~

3~ 7~?7ever, upon studying the di9clogure in the aforelnentloned ~atent
No. 4,030,351 one wlll note that tl~ere i9 a dedicated computer devoted
to just one test 9tand, In clle ~roductlon t~sting of carburetors, a com
puter 19 normally u8~d to conl:rol as many as sixteen (16) or more test
stand simultaneou~ly,
When you clo.qe a test loop w~th a computer in thls fnsllion, you
restrict Ihe computer'~ ability to perform any other tasks efficiently,
thereby slowing the entire process. It wa9 for thi3 reason that an exten-
sion of the laboratory test stand concept to the production line was lm~rac- -
tlcal. Also, it vllould be prohlbitively expenslye to have a dedicated com-
puter for enc'n production ~est sta.nd wh~n the qu~ntity of production type
test seand is considered, Thus, ~llile l~boratory ~ccuracy could be
obtaincd, th~ obtaininà of it at production rates provided major obstacles.
rhus, vre needed to find a novel way to llave accuracy without a dedic~ted
computer.
.
By looking at conventlonal three-mode controllers presently on
the market, such as the Model No. 52H-SE made by The Foxboro Corn-
pany of Fox~oro, Massachusett~ in an àttelnpt tO still use a conventlon~l
con~roller for accuracy, but to get away from the need for a computer, it
wa~ very qulckly found that '~ecause of certain operacional characteristics
such controllers were not useahle. A major consideration was that such
controllers do not haYe a deflnite dead l~and. In other ~ords, even thou~h
the process controller ~ould operate tlle carburetor to get the throttl~
plate to the des~red position, one could noc automatically and economlcally
stop the action of the process controller at that point, ~ thus one u~ould
have a continuous huntin~ situation around the desired set polnt, and one
could not ~et a scal~le process,

Further. there W~5 nota slngle proces~ contro~er onthe mar-
ket ~hatcontrolled proces~ operacin~ devices of all threetypesth~t~Yere
required, namely~he DC s~eppin~ mo~or, the AC synchronous motor and
che pn~umatic or hydraulictype positioner. ~hls obviously then could
not be a~ensible solution, since~he util~zatlon ofthe avalla~le controller~
would notproduce a proce6s con~roll~r capable ofh.~n~ling allthe sltuatlons
whlch ~re encountered. Further, the sLandard co~troller~foundto ke
avail~ble were capable ofcontrollinJproce~es o~y over a relatively nar-
row ~ange and did nothave proportional, rate, and re~etfunctions wllioh
were ~uitable to the proces~es wh~ch hadto be controlledinthe pro~uction
testina of carburetors.
Abandoning the old three mode controllers previously
used and developing our own novel controller which controls a
process as a function of ~he difference of, and rate of change
between, a desired value and a current state of the process,
and includes a deadband feature, we have developed a control-
ler which gives laboratory results on a production line basis.
During the time when development was going on in the
invention of our novel single-state four-mode controller, even
more stringent requirements were placed on Applicants assignee
to determine or make a machine which would test carburetors
on a laboratory basis more rapidly then possible with the
optimized rate, reset and proportional control discussed early
in the present application. We thus had to find a way to
speed up laboratory testing of carburetors also, for example,
make their testing time 1/2 of that previously obtainable. To
do this however, proved to be quite a problem.
.
Therefore, we were forced to reevaluate the systems
used and previously described in our patent No. 3,517~552,
3,524,344, 3,851,523, 3,896,670, 3,975,953 and 4,030,351.
~ Ç~ ~

~ t7~7
In the earliest of the patents as discussed there was
either a one speed throttle drive which could operate in either
direction or a two speed throttle drive with a deadband which
of necessity had to have a circui~ry designed so that they
were not driven too ~ast because of a coasting problem inherent
in the drive motor. You had to have a rather wide deadband
also to stop the motor and hope that one would not coast out of
the deadband or the system would go into a hunting condition
which when applied to any process being controlled would
greatly decrease the ability to properly test the part and also
greatly increase the time for going from one test point to
another.
The second type of system we looked at was the one
wherein the rate of ro~a~ion of the throttle pla~e of ~he car-
buretor, or by analogy a process device) was proportional to
~he difference between the actual and the desired process set-
ting. We found that we could no~ speed up this system because
of the same overshoot problem just mentioned and the fact that
because this system was now several years old all of the cir-
cuitry which was designed for it used the standard type motor
drives which could not be changed. We found also that if we
went to a high speed motor drive we would again cause the over-
shoot problem and end up in a hunting situation. We therefore
had to abandon the idea of speeding up the proportional control
type of system.
,.

We next looked at the system descr;bed in our paten-
~No. 4,030,351 which involved the optimizing of rate, reset and
proportion in going rom one test point to another.
Theoretically we though~ this would give us our solu-
tion if we could optimize ~he values in combination with the
saturation of the circui~ry previously described. However, we
found an unexpected problem since at the time the circuitry
was designed for the system which optimized rate, reset and
proportion it was primarily designed for use a~ a constant
altitude near sea level a~ the different test points wherein
the optimization of the values was relatively easy.
However, when we ~ried to apply such technique to the
testing of a carburetor under current reglations which require
tha~ the carburetor be -tested regularly at various altitudes,
we found the operation o the stand and the optimizing of the
ralues became unwieldy to ~he point of being uneconomical to
achieve even when a computer was used to aid in operation of
the system.
.
Also, it must be remembered that our process controller
is intended to be used to control many different process at
many different desired values and it was found tha~ whether a
carburetor throt~le is being controlled or a valve which may
be used in a manifold vacuum or a hood pressure system, there
was now a need to optimize the values for many different de-
sired values which, while possible, was no longer economical.

l~-17~6-5~
Thus we abandoned the idea or trying to obtain a faster test
using a test device which works on the idea of optimizing the
values oE rate, reset and proportion.
Having tried all these existing ways to solve the
problem of moving from one test point to another faster and
having failed, we took a serious look at our basic premise of
trying to avoid a process overshoot and decided to try a new
approach of fir.st intentionally causing both the process device
and the process itselE to overshoot to get -the process in the
approximate position of the correct value very fast, second
having the controller reverse direction at a predetermined rapid
speed to again cause the process to approach the new set point,
and third controlling the process as previously described until
the process was within a preselected deadband.
The e~Eect oE this will be described hereina~ter by
reEerence to a graph of time versus the process correlate signal
and process device position. Upon looking at a graph of the
process correlate signal in relation to the process device
position and by viewing this for what we choose to call the
three-states of operation one can see that if one supplies a new
desired value, one will cause th~ circuitry to saturate, as will
be described hereinafter, and the process device will start
moving rapidly with the process correlate signal following suit.
It is to be noted that the process device is continued to be
moved until the process correlate signal changes in polarity
_ g _
f
,, ",

77
D18~ 6-54
which means the process has reached the desired value for the
Eirst time completing what we shall term state one. The
circuitry then enters what is called state two wherein the
direction ot process device movelnent is reversed. The process
device is operated in this reverse direction rapidly while the
rate oE change o~ the error signal between the desired value and
process correlate signals is now watched in addition to the error
signal itself. It should be noted that the speed of such rapid
movement is chosen by consideration of the response time of the
process, and thus oE the process correlate signal.
~ hen the summation of the error signal and the rate of
change signal changes polarity, the circuitry en-ters state three
which is a return to the OQeratiOn previously described in regard
to the single-state four-mode controller. The effect on the
test time by using this new method of operation will be
graphically illustrated hereinafter by cornpariny -the operation
time of a strictly proportional circuit, -the operation time of
the single-state four-mode controller just described, and the
operation time of the three-state four-mode controller would
typically take to move to a certain set point. The savings in
time in using the three-state four-mode controller is very
significant in view of the capital investment which must be made
in test equipment today and the ever increasing need for more and
more laboratory type tests to meet current regulations.
- 10 -

~4~m
Before proceeding to the detailed operation of the three-state
four-mode process controller, a brie~ discussion of the defini-
tion o the states and modes is in order. State one consists
of a predetermined rapid, constant speed process device move-
ment which continues un~il the error between the feedback
signal and the desire~ value changes polarity. S~ate two con-
sists of a predetermined rapid, constant speed process device
movement in the reverse direction which continu~ until the
summation of the error between the feedbac~ signal and the de-
sired value signal and the rate o change of said error changes
polarity. State three consists of ~he our-mode operation, in
which the four modes are proportion, rate, minimum speed, and
deadband as previously described.
Thus, one the objects of ~he present invention is to
provide a new and improved process controller capable of
providing laboratory accuracy at production process speed.
Another object of ~he presen~ invention i5 to provide
a controller of the above nature having a definite deadband
capability.
- Another object of the present invention is to provide a process controller which is capable o controlling DC
stepping motor type operators, DC Servo motor operators, AC
synchronous operators) and pn~umatic or hydraulic positioners.
A further object of the present invention is to pro-
vide a process controller haYing a wide range capability.

~6r~7
A fu~ther object of the present invention is to pro-
vide an improved single-state Eour-mode process controller
having rate, reset and proportional types o ac~ion which
will quickly and accurately reach a value within a deadband
range of the desired value and turn itself off, thus eliminat-
ing any ~unting cond;tion.
A further object of the present invention is ~o pro-
vide a four-mod0 process controller of the above nature which
is capable o manual or automatic control.
A still further object of the present invention is to
make an improved process con~roller which can easily set
processes to a multitude of different conditions for use in
setting different process conditions and can be directed to
do so by an automation device.
A further object of the present invention is to pro-
Yide a process con~roller of ~he above nature which is capable
of controlling manifold vacuum across a carburetor during a
carburetor test cycle.
Another object of ~he present invention is to provide
a production type process controller capable o~ obtaining
labora~ory accuracy while controlling pressure inside a
carburetor ~est hood.

Another object of the present invention is to provide
a production type process controller capable of controlling
the pressure of a liquid in a conduit in a quick and accurate
manner.
Another ob~ect Or the present invention is to provide
a process controller o the above-described nature which is
suitable for con~rolling air flow through a carburetor.
Another object of the present invention is to provide
a production type process controller which is reliable and
relatively inexpensive to manufacture.
~ .
Another object of the present invention is to provide
a ~wo-directional switched driver capable of controlling the
operation of any two-directional device, such as an AC syn-
chronous motor.
. .
A still further objec~ of the present invention is to
provide a new and improved three-state four-mode process con-
troller for laboratory use which will perform laboratory car-
buretor tests at rates much faster than previously possible.
A still further object of the present invention is to
provide a laboratory type carburetor ~est facility in which
movements from one test point to another test point are made
very rapidly by the use of rate, reset, proportional, and
deadband control.

A still ~urther object of the present invention is to
provide a laboratory carburetor test stand o the foregoing
nature in which the device controlling the process in question
is moved rapidly until the error signal representing the error
in current state of the process changes polarity, and then the
device is reversed in dii~ec~ion and moved rapidly until the
summation of the error signal representing the error in cur-
rent state of the process and the rate of change of said error
signal changes polarity after which said system will operate
in the normal manner using the combination of the rate, reset
:~ and proportional types of action until the signal is brought
within the deadband range at which time the movement of the
process device will stop.
" ` .~
~ Further objects and advantages of this invention will
:~ be apparsnt from the following description and appended
claims, reference being had to the accompanying drawings form-
ing a part of this specification, wherein like reference char-
acters designate corresponding parts in the several views.
~ .
~ Figure 1 is a general diagrammatic view of a closed-
~ loop process embodying a process controller utilizing the con-
struction of our invention.
.
Figure 2 is a diagrammatic view similar in part to
that shown in Figure 1, but showing a closed-loop process
which has to repeatedly be se~ to many conditions and thus
embodies an automation device in connection with our improved
process controller.

77
Figure 3 is a view of a closed-loop process embodying a process
controller utilizlng the construction of our pre~ent invehtion and adapted
to be operated manually.
Figure 4a l~ a diagramm~tic view of n manlfold vacuum control
process which may be controlled utilizing a process controller ernbod)~ing
the constructlon of our present inventlon.
Flgure 4b is a diagrammatic vlew of a hood pressure control pro-
cess ~hich may be controlled utilizing a~ pxocess controller embodying the
con~truction of our present inventlon.
Figure 4c is a diagrammatic vlew of a fuel pressure control pro-
cess ~Ahich may be controlled utillzln~g a process controller embodying the
constructioll of our present invention,
Figure 4d show~ an air flow measurement system which may
embody the process controller whlch utilizes the construction of our pres-
ent lnvent~on to control air flow.
Figure 4e shows an air flow ~neasurement system slmllar to that
shown ln ~igure 4d, but uslng sonic flow devices, utillzing ~he proces3
cor.tr~l1er embodylng ~he construction of our presenc invention~
Flgure 4f i9 a view similar to ~hat shown in Figure 4e, but hav-
ing the air flow measurement system operating ln a controlled environ-
ment wherein a differential pressure transducer may be used to form the
feedback signal devlce in place of the absolute pressure transducer.
~"~_

7'77
Figure 5 is a schematic dlagram of one embodimen~ of the dif~er-
cntlal input clrcui~ embodied ln the process controller utl1izing the con-
structlon of our present invention,
Flgure 6 i9 a schematic dlngraln of one embodirnent of a correc-
tlve action circuit used ln the process controller embodying the cons truc -
tion of our present invention.
Figure 7 is a schematlc view of another embodlment of a correc-
tlve actlon clrcult which rnay be used in our novel process controller.
Figllre 8 shows another embodlment of a correctlve actlon cir-
cult which m~y be ~Ised in our novel process controller.
Flgure 9 i8 a schematic diagram of the valld range check circuit
embodied in the construction of our present lnvention.
Flgure 1018 a 9chematic diagram of the error and ra~e ampllfier
clrcult used in the construction of our present invention~
Figure 11 ls a schematic diagram of an embodiment of a scaling
and meter protectlon circuit embodied in the construc~lon of our presenc
lnvention~
Figure 12 ls a ~chematlc diagram of a bu~fer-scaler which may
~e em~odied in the construction of our present invention.
Figure 13 shows a summing ampllfier eml~odied in the construc-
tion of our prese~t inventlon.
16

77
Figure 14 1~ a schematlc di~gram showing sn embodimen~ of an
lntegrator a~ u~ed ~n th~ construction of our present inventlon.
Flgure 15 iB a schematic dlagr~m of a ~umming integrator
which may be used in the construction of our present inventlon.
Flgure 16 ls a schematlc diagram of an absolute value circult
whlch may be embodied in the construccion of our present inventlon.
Figure 17 is a schematic diagram of a two-directlonal s~ltched
drlver whlch may be util~zed ln the construction o~ our present invention
when a reversible AC synchronous mocor or oth~r reversible devlces
are to be utlllzed to control a process with our process controller.
~ .
Flgure 18 ~6 a schen atic diagram of a reverslble AC synchro-
nous motor, ~hlch may be the operator controlled by our improved
process controller.
. ' .
Figure 19 i8 a schematic diagram of a reversib!e DC motor
whose dlrectlon 18 controlled by a pair of relay contac~a connected to
opposite polaritles.
:
Figure 20 i9 a schematic diagram showlng how a pair of sole-
- no~ds may be connected.
F~gure 21 is a dlagrammatic,view showin~ how the solenoids of
~lgure 20 ~nay be connected ~o operate a pneumatic or hydraullc cylinder.
., .
-17 -

~ 7 7
Figure 22 is similar to ~igure 1 i~, that it is a
general diagrammatic view of a closed-loop process, but in
this case embodying a three-state four-mode process control-
ler utilizing the construction of the present invention.
Figure 23 is a diagrammatic view similar in part to
that shown in Figure 22, but showing a closed loop process
which has to repeatedly be set to many conditions, and which
thus embodies an automation device in connection with the
three-state four-mode process controller.
Figure 24 is a view of a closed-loop procesi embody-
ing a three-state four-mode process controller embodying the
construction of our present invention and adapted to be oper-
ated manually.

~ ~7~P7
Figure 25 is similar to Figure 22 but in this case
utilizies a process speed improvement device of a type to be
described hereinafter to enable the entire process to move
~rom one position to another at an increased rate of speed.
Figure 26 is sim;lar in part to Pigure 4b, and shows
a hood pressure control system of the type which may embody
the three-state four-mode controller having the construction
of our present invention, and utilizing a process speed
improvement device.
Figure 27 is an overall diagrammatic view of a test
system which may be constructed utilizing the controllers of
the present invention, and showing as subsystems thereof an
air flow measurement and control system, a manifold vacuum
measurement and control system, and a hood pressure measure-
ment and control system. The hood pressure~ manifold vacuum,
and air flow measurement and controls and system utilize a
three-state four-mode controller embodying the construction
of the present invention which will be described in detail
below.
~. '
Figure 28 is similar to Figure 27 but includes the
use of a process speed improvement device in the hood pres-
sure measurement and control system.
Figure 29 is a view similar to Figure 27 but utiliz-
ing a computer system for automatically testing a carburetor
in the laboratory at several test points.
,: ,
,/ ~i'

~6m
Figure 30 is similar in large part to Figure 29 but
using the process speed improvement device to more rapidly
test the carburetor in the laboratory under many test points.
Figure ~1 is a view similar to Figure 30, but showing
an air flow measurement system, and utilizing the computer for
controlling the carburetor throttle plate rather than having
the subsystem itself controlling it.
Figure 32 is similar $o Figure 5, but showing a three-
state differential input circui~ including a three-s~ate error
and rate amplifier circuit as utilized in the three-sta~e
four-mode process controller.
Figure 33 is a graphical representation showing ~he
three different states utilized by our three-state four-mode
process controller and the values of the process correlate
signal and the process device position as a function of time.
Figure 34 is a graphical representation of time versus
process correlate signal showing the comparative time a
process controller will ~ake to move from an old set point to
a new set point using various process controllers. This
figure shows relative times ~or systems using a three-state
four~mode controller, a single-sta~e our-mode con~roller,
and a rate plus proportion type controlO
- 2G~>-

;777
Figure 35 is a ~iew similar to Figure 10 but showing
the three-state error and rate amplifier circuit which is
used in the three-state four-mode controller.
It shoul~ be understood that the present invention is
not limited in its application to the details of construction
and arrangement of parts illustrated in the accompanying
drawings, since the invention is capable of other embodiments
and of being practiced or carried out in various ways within
the scope of the claims. Also, it is to be understood that
the phraseology and terminology employed herein is for the
purpose of description and not of limitation.
~21-

Tllere is sllown in ~igure l a typlcal use of our lmproved process
controller, generally deslgnated by the numeral 40. The process con-
troller is supplied with a voltage reference 1ndicat1n~ a desired value from
a d~sired sett1ng device 41 whlch causes the controller to supply a s1gnal
to the drlver 43 which, in turn, 5upplie~ a process 1llput s~nal to the
proces9 generally de~ignated by the numeral 4~ nt the connectlon labeled
48. Since tllis ls a clos~d-loop system we ~re concerned w~h, the process
4~ will tllen supply a process corr~la~e sl~nal 49 ~ndicating the current
s~ate o~ the process. If the correlate slgnal i9 a volta~e signal useable by
the process controller generally desl~natecl 40, lt may be directly ~upplied
there~o. If, however, the correlate signal is not dlrectly compatible, a
feedback sl~nal device 42 is needed to convert the s~gnal into one useable
by the controller. For exarllple, lf the process correlate si~llal ~9 is
pneurllatic ~n natux~, the ~eedback 6ignal devlce may take th~ ~orm of a
pressu~e ~ransducer.
Since ~he rneans for conYerting the~e si~nals are ~vell known ln
the art, and the types of conver~ions needed are so numerou~, lt 1~
believed not practicable tO d~scrl~e all ~he varioLis poss1bilities ln the
present applicatlon. I~ suffices to say ~l at one sl;illed ln the art would be
~ble to provlde a proper feedb~ck ~ignaldev~ce 42.
It should be understood that the process 44 under
control generally consists of a process measurement device 47
which is used to measure the current state of the process, a
process device 46 which is used to change the current state
of the process, and an operator 45 which is used to change
the process device.
-22-

7~7
V~hile Flgure l has shown a generaliæd dia~rammatic view of
a closed-loop 6ystem ernbodylng our proces~ controllex 40, Figure 2
shows an embodiment of our lnven~lon whexe it is deslrcd to automatically
operate at a varlety of desired settings, such as to test over many test
polnts of a devlce ~uch as a carburetor or the like, where one may test
over as many as 30 polnts. Scme modlfication ls needed for this sltuatlon
over the gener~liæd ver~ion because you would need a new deslred value
from ~he de~lred settlng device 41 for eacll test point, While these could
l~e set manually, as will be discussed below ln relatlon to Flgure 3, it is
much easler to have an automatlon device 54 which will automatically
change the desired value for the neXt conditlon upon completion of the test
at the present test polnt, It is also pos~lble, as ~hown by the dotted line
ln Figuxe 2, tO tle the outpu~ from the feedback signal device 42 or the
process correlate signal 49 to the automation device 54. ~Fhls may be
deslred to conflrm tllac the particular condition at Y~1hich the proceqs has
arrived is indeed the desirèd condltion before the a~ltomation devlce 54
takes fur~her action.
A~ shoYvn in Figllre 3, a manual system i~ posslble uslng our
invention where the particular design requiremen~s for the system permit
it, or uhere economy dictate~ such a system. I,1 ~hiB case a po~entiometer
55 could actually be the desired ~ettlng devlce 41. I
It ~hould be understood that ~here rnay be some conve~lon or sig-
nal conditioning necessary of the slgnal from the feedback si~;nal device and
of the actual si~nal from ~he desired seeting device 41, which 19 ~et elther
manually or by the automation devlce 54 before the signals can be used by
_,~

t777
the process controller 40, ~,ain the number of possibilities of conver-
~lon and signal condltionin~ nleans are nurnerous nnd so well knou~n in
the art, that it is not deemed necessary to describe them further herein.
As ,qn exalllple of processes ~ lch can utlll~e oux lmprovecl
process controller, ~llere are shown ln Figures 4a to 4f si~c di~ferent
exalnplea. Referrlncr speclfically to Fi~ure 4a, the process 44 in this
example i~ one wherein the manifold vacuum acros~ the carburetor 56
must be precise1y controlled, and must be able to ~e set to different test
condltlon~ rapidly. In ~lls in9tance the carburetor 56 is mounted on a
rlser 57 ln any ~ultable manner inside the hood 59. In order to control ~he
manlfold vacuum across the carburetor, it l~ of course first necessary to
kno~v ~rhat the actual manifold vacuutn i~ ~t any glvell moment, For this
purpose, a diferential pressure transducer 47a becomes the process
measuremen~ device, and is capable of givln~ a process correlate signal
49 as an output. Such a differentia1 pressure transducer, which may be
such a~ the 1151 DP serles manufactured by Rosemount Englneerlng Co. of
Minneapolis, Minnesota has a high pre~ure input 60 connected to sense
the pre~sure above ~he c~rl)uretor under ~he hood S9~ ancl a low pressure
input 58 connected lll the throa~ o~ tlle carl~uretor riser ~7 to ~ense clle
pressu~c beneath tlle c~rburetor~ By metllods ~ell lcnov.~n ln the art ~uch
differential pressure traIlsducer tl1en produces a process correlate si~nal
~9 contln1l0us1y related.to the pressure drop across the carburetor ~t any
glYen po~nt, which ls commonly known as the manlfold vncuum.
Now referrin~, baclc ~o any one of Figures 1, 2 or 3, SUCIl process
correlate signal would be fed throu~h a feedl~ack slgnal clevlce 42, lf nec-

,i
~ 3L4~777esBary, and then fed into the proce~ controller 40. The proce~ con-
troller would compare the proce~ coxrela~e slgnal with 8 deaired value
and, if necessary, provide a corrective action ~ignal to the driver 43,
wh~c11 the drlver would then conv~rt ~n a manner to be dc9cr1bed hereln-
below, to a proce9s input ~ignal 48 capable of drivlng the operator 457
This then clo~es the loop and thi9 operation w1ll contlnually take place
untll the operator 45 causes ~he proces~ device 46 to mov~ to a pos1t10n
such thae the proces9 change~ resulting in a change to the proce~s
measurement device 47 cau51ng the p~oces~ correlate slgnal to become
~table and to corre~pond tO tlle desired cetting 41. At this polnt the
process will have stab1l1zed at the desired value. Once the procecs i~
~table and at the desired value~ the process controller rema1ns active,
continuou~ly xepeating the compari~on and correctlon process. lJpon
a process change for any reason or a new desired value, further cor-
rection i9 made until the proce~s i9 aga1n at the deslred value, It can be
seen that thi~ operatlon holds true whether the 6ystem 1~ the generalized
Yers10n shown ~n Figure l, the automated version a~ shown in F1gure 2,
or the manual version a~ shown ln Figure 3.
Referring again to Figure 4a, the operator 45 is in
the form of a valve operator 45a. This then closes the loop
and this operation will continually take place until the valve
operator 45a causes the process device 46, which in this case
is a valve 46a, to move to a position such that the process
changes result in a change to the differential pressure trans-
ducer 47a causing the process correlate signal to become
stable and to correspond to the desired value signal. At this
25 -

7"~7
point the process will have stabili~ed at the desired value.
Once the proc~ss is stable and at the desired value within the
deadband range, ~he process controller remains active, con-
tinuously repeating the comparison and correction process.
Upon a process change for any reason or a new desired value,
further correction is made until the process is again stable
at the desired value wi~hin the s~lected deadband range. It
can be seen that this operation holds true whether the system
ls the generalized version shown in Figure 1, the automated
version as shown in Figure 2, or the manual Yersion as shown
in Figure 3.
Another example of a process ~hich can be controUed by our
lmproved proces~ controller i8 that shown In Figure 4b ~!here ~t is desired
to accurately control the pressure inside the hood 59~ In order to control
such pressure one must mea~ure ~he hood pressure, and thls ~s done by an
absolute pressure transducer 47b which may be such a~ the 1332 serles manu-
factured by Rosemount Englneering Co. of Mlnneapolis, Mlnnesota. In a
manner well known ln ~he art, ~aid absolute pre~sure transducer produce~
~ ~? G

7~7
a process correl~te signal 49 whlch, in a manner slmilar to that Ju~t
descrlbed, is fed through a feedback signal devlce 42, lf necessary,
and then fed lnto the proce~s controller 40.
~ previous1y descri~ed, the proces~ correlate signal 4~ would
J~ compared ln a mann~r ~hown ln Flgure~ 1 to 3 w1th a signal from the
desired setting device 41, and lf a difference exlsts between the actual
state of the process and the desired sLate of the proceas, the process
controller would then supply the necessary 6ignal to the drlver 43 to drive
the operator 4S, which ln thiq case i8 a valve operator 45b drivlng the
process device whlch ls ln the form OI a valve q6b. Agaln the new process
correlate ~1gnal 49 would be supplled tC) the con~roller, colnpared to the
s1gnal from the desired ~1gnal devlce ~l, and,1f necessary, si~nals would
be given to the dri~er 43 whlch would a~aln produce a new process lnput
slgnal 48, w~th the proces~ continually repeating icself untll the de3ired
~ralue i~ reached,
Referring to Figure 4c there i8 ~hown a process 44 adapted to "~
control the pressure of the fuel being supplied to a carburetor a~ other
llke devlce. In this case~ slmllar to that previously described, the car-
buretor 56.would be mounted on a riser 57 inside the hood 59, with fuel
from the fuel source ~not ~hown) passin~ through a first condu1t 64 through
a proce~ device 46 ~n the form of a valve 46c through a second condult 65
and into the carburetor 56. A process lnput signal 48 i9 supplied to the
valve operator 45c which operates the valve 46c to perform the actual
~nction of controlling the pressu:re withln the second conduit 65. I~ should
be understood that carburetors are also tested wlthout use of hoods, and
-~7-

i'7~
~he pressure of ~he fuel suppl~ed to the carburetor may be controlled by
our lmproved process controller ln such a system wlthout a hood.
To obtaln a measurement of the pressure ln the conduit 65, a
dlfferentlal pressure transducer 47c is used as tl~e process n~easurement
devlce~ Connections to the high pressure lnput 60 and the low pre~sure
lnput 58 ena~le the dlfferentlal pressure transducer 47c to determine the
pressure in the system at any given time and supply the proces~ corre~
late slgnal 49 to the process controller 40 through a feedback ~ignal
devlce 42, if needed, Agaln the comparison and correction proces~ wlll
take place ln a manner previous1y descrlbed untll ~he process i8 at the
deslred valu~ ~wi~hln the dead band range of the process con~roller, The
comparison process contlnues to occur whl1e th~ proce~s ls wlthln the
dea~ band ran~e untll the process goes c)utside of the dead band Yihether
due to a process change or a change lrl the desired value, ~t ~hl~ time,
the c~rrectlon proces~ again occur~ until the proce~ is again at the
desired value.
In carburetor testing it i9 al80 necessary to measure the air
~ow to the carl>u~etor, which ~n thia case is control1ed by the carburetor
itse1f. Thu~, the carburetor prevlously referred to under the numeral 56
becomes the process dev~ce and is now referred to by the numera1 46d.
In order to mea~ure the alr flow ~hrough ~he carburetor, a hood 59 ls
provided whlch has an outle~ 62 connected tO a vacuum souxce, and an
lnlet 63 connected to an air flow measurement system 47d, whlch may be
as subsonlc nozzles or laminar flow tubes. The quantity of alr flowlng
through the carbure~or 46d then is controlled by the movements OI the

. ( ~ (
throttle plate~ which 1~ cont~olled by the tl~ro~tle operator 45d, The
throttle operator 45d 1~ controlled by the process input 4Ignal 48.
To arrive at A de3ired air f~o~v through the carburetor, i~ i8
necessary to knov~ tlle ~ir flow pre~ent ln the ~ystem at any time. In
thi6 c~se, the alr flow "lea6urement system ~iU provide a pressure cor-
relate 9ign~149 In the forrn o~ a dlfferentlal pre~5ure si~nal ~hich will
be supplied to the feedback ~ignal devlce 42, which now takes the form of
a dlfferential pressure tr~nsducer 42d~ l his~ ln turn3 will supply the
Rignal to the process controller relatin~ to the current air llow conditions
through the carburetor 46d. In a m~nner ~imilar to that previously
de~cribed, the co;nparison and correction operations wlll take place untll
the desired vllue ~ithin dead band llmits ls reached.
When it 1~ desired to have a sonic nlr flow 1neasurement system
u~lng crltical venturl meters or variable area critical ~enturi me~er~,
tlle sy~te~ shown in Figures4e arld 4f may be the ones controlled by our
process controller. Referrlng to Figure 4e, it i9 actually the carburetor
which i~ the p1ocess control device as ln Figure 4cl, and 1~ 1B~ therefore,
no~ labf~led 4Ge r~ther than 56. The turnlng of the carburetor throttle
plate by the throttle opexator 45e controls the amount o~ air passing
through the car~uretor,
Since sonic lir flow Mensurement is be~ng used, whereln aLr
flow i8 baslcally proportlonal to the ab~olute pressure, ~he carbur~tor
hood S~ previou~l y described ls not reguired,but may be used, The car-
buretor 46e will be rnounted on the riser S7 as prevlously descri~ed,
-~7~

7'77
~he process lnput signal 48 drives the throttle operator while the pres-
sure signal from the air flow measurement system 47e i9 the proces~
correla~e slgnal. Sald process correlate slgnal 49 i~ supplied through
the conduit 61 to the absolute pressure transducer 42e. The process
correlate slgnal 49 ls transformed lnto a signal compatlble with the
process controller by the feedb~ck ~ignal device 42 ln the form of the
absolute pressure txansducer 42e. A~ain, the signal, ln a manner ~iml-
lar to that previously described, is compared with a de~ired value ~ignal
from a desired value setting device and, if necessary, the process con-
troller supplies a ~lgnal to ~he driver 43 which, iff turn, suppl1es a
proce~s input signal 48 to the operator 45e. ~h~( comparison and cor-
rectlon process will continue until the process correlate signEll correa-
ponds to the desired settlng~ thus ~etting the air flow through the car-
buretor 46e to the desired v~lue within ~ead band limits of the process
controller.
Another ~ystem 4~ ~or setting the air flow through the carbure-
tor uslng the 80nic flow devices i8 shown in Figure 4f. In this case, the
throttle operator 45f, the ca~burecor 46~, and the carburetor riser 57
may be the ~me as those lndicated by numerals 45e, 46e, and 57, shown
ln Flgure 4c, However, to ut11ize ~ transducer wlth ~ sm~ller spaa, the
differentisl pressure transducer 42f may be used instead s~f the absolute
pressure transducer 42e ~o form the feedback slgnal devlce, In thls cas
the measurement of air flow i8 taking place as a function of mQnifold
vacuum because when the process 44 is being performed ln a controlled
atmospheric room, manifold ~racuum relates to absolute pressure and,
therefore, alr flow is a function of the manifold vacuum. Thus, the process

7~7
correlate slgnal ls tl~e dlfferentlal pressure signal 49, and tl~i~ would be
supplled to the dlfferent i~l pressure transducer 42f, The sl~nal from the
feedback slgnal devlce, in this case a dlfferentlal pressure transducer
42~, w~uld be used In a manner deYcrlbed lmrnedlately above to produce
any changes necessary in the process Input slgnal 48 until the process
inpuc ~Ignal ~8 corresponds to the proces~ correla~e signal 49 and the
process 18 at the deslred value witnln dead band limits of the proce6s con-
troller.
The descrip~ion thus far has dealt substantially wlth lllu~tratlons
of a general nature showing varlous closed-loop p/rocesses embodylng our
inventlon and the type~ of proees~es they can control, and has not dealt
wlth any detallecl descriptlon of the operation of t1le process controller
itself, or of lts novel features over those controllers known in the art.
;
To more fully understand the novelty and operatlon oE our Inven-
tion, It i9 eo be noted that the proce~s controller 40 shown In Flgures 1,
2 and 3 consists of two portions, the dlf~eren~ial lnput clrcult 67 ~nd the
corrective actlon clrcult 68r In general, the dlfferentlal input circult
compare3 the proce~s cci~rel-~ ~ignal with the deslred value ~Ignal from
~he desired settLn~ devlce, finds the actual error dlf~erence between the
two slgnals (static~, finds ~he rate of change (dynamlc) between the two
signals, ~ums them algebra~cally, and then provlde~ an OUtpLlC signal to
be u9ed by the correc~lve action clrcult 68 to control the drlver 43, a~
necessary. If the deslred value 18 within the 5et poln~s 72 and 73~ the
error and rate amp1iflcation circuit 70 wlll operate normally, resultlng
~n the approprlate correction slgnal belng supplled ~o the correc-

f --
'7~
tive actlon clrcult 68. However, if the desired value i3 outslde the validrange set polnts, thls w1ll cause the error and rate ampllfication clrcuit
to become saturated and go to a full plus or full mlnus saturated condition
depending on whether the deslred value VJa8 outside the hlgh llmlt set
polnt 72 or the low limit set polnt 73. Thl~, ln turn? wlll ultimately
cause the process devlce 46 to rapldly go to one extreme or another, for
example~ fully opened or fully closed, and stay there untll eome further
slgnal~ are received Irom the circui~y.
It should be understood tha~ the proces~ 18 generally one of a
dynamlc nature, and the process controller is attempting to ol)tain a ~table
stat~c condltlon. If the correction ~1gnal from tlle error and rate ampllfier
circuit 70 ls wlthin dead band limits, the process controller 40 provides
a ~tatic output signal and ~he control remains held until an upse~ or chan~e
in the process causes the proce~s to go outside the dead band llmit8. The
pIOCe9E3 will be considered to be w~thin the dead band llmi~s when said
correc~ion ~ignal i9 essentially at zero value9 which may be when the rate
of change is equal ln value to the error ~l~nal, but oppo~ite in polar~ty,
or when the rate of change i8 at a zero v.~lue.
Referring to Figure S; the feedback arld the desired valu~ signals
are fed to both the error and rate amplif1er circuit 70 al~d ~o the scaling
and me~er protection clrcult 71. Addlti~nally, the desired v~lue signal ls
fed to the valld range check clrcuit 79. The purpose of the error and rate
amplifler circult i9 to algebraicall3~ sum the actual difference between the
feedback and the desired value signal, whlch is a s~atic error, and the
rete of change of the feedback ~lgnal wi~h re~pect ~o the deslred value sig-
..,~,~,

slal, whlch is a dynamic error. Additlonally, ~n order to prot~ct theproce~s equipmentj A valld range check circuit 69 i8 yrovided Thi3 i8
necessary because in some embodlments of our inventlon, the stepping
motors used can easily darnage the equipm~nt bein~r tested due to the
~notor characteristlcY. Ag 18 well kno~n ln the art ~se~ Desl~n En~lneer's
Gulde to QC Stepplng ~otors by Superior Electric Co~npany, 13rlstol,
Connectlcut) at very hl~h speed~, stepplng ~notors have very low torgue.
However, ~t the low speed~ the torqu~ i8 very hi~h. I hus, in certaln
types of tests, for exasnple a carl~ur~tor test where the stepping motor i8
turnlng the clrburetor tllrott1e plate, when the deslred value i9 OU~ of
ran~e, ~n undesirable condltlon could occur, namely that the carburetor
throttle plate could become fully closed or fully opened witll the s~eppin~
motor turnlng 810wly with large torque. The c~xburetor could easlly
become damaaed, or the mechanlca1 connectlon ~tween the stepping
rnotor and the carburetor could become damaged,
~ o prevent this,- the valid range check circult 69 compares the
desired value agalnst the higll limit set poin~ 7~ and the low llmit set
point 73, as shown in F~gure 9. If ~he desired value is v/ithin the valld
r~nge set poin~s, tlle vali~ range check clrcuit 69 will cau~e the error
and r~te amplifier circult 70 ~o operate in itS r~orrnal mode supplylng the
correction signal to the correctlve actioll circult 6~ Howev~r, lf the
deslred value ls outslde the valld range set pOilltS, the valld range check
circult will act in ~ malmer to cause the stepplng motor tO operate ~t its
maximum spee~ and drive the process device to its fu11y closed or fu11y
opened posltion As prevlous1y mentloned, at Eull spee~ steppln~ motors
, . ,
-33 -

7'77
have n very low torqu~, ~o in thls ca~e when the proce~s device r~ache~
its fully opened or fully ~lo~ed positlon, the stepping motor wlll slmp1y
9t~ , causlng the proce~s dcvlce 46 to cease further ~djustment. Upon \~
becomln~ aware of thi~ conditlon, the oper~t~ng per~onnel can take the
necessary action to correct this situ;-tion.
Typlc~lly, in a proces~ control circuit th~re ls provlded a
deviatlon meter to lndlcate the relation3hlp between the current condi~lon
of the proce~ and the deslred 6et point. Since these process ranges are
usually rather large, and the clesired meter rallge i~ relatively ~mall,
it is nece~sary to provide a means of ~cAlins the available error signal to
a signal useable ~y the meter~ lt i~ a1so desirab1e to protect ~he meter
~rom an over10ad conditiorl should the process error exceed the range,
This i~ done by the ~caling and rneter pro~ectlon clrcult.
. .
A detailed clescription of the operation and componen~s of the
valid range check clrcuit, error and rate amplifier circui~, and ~calin~
and meter protectlon circuit can be found ln Figures 9, ~ and ll, re~pec-
tiYely,
~- In Figure ~, the valid ran~e ches:k cLrcuit ~ operate3 by COIl-
necting a hl~gl1 limit se~ polnt 72 ~o the hi~h limit compar3to~ 7~ and the
low 11mit 6et point 73 tO tbe low liinlt compar~tor 75. ~ the same tlme
the desired value ~ignal l~ supplied to both comparcator~, which can be
8uch as Mode1 8311 made by Ana10g Devic~6, Inc. of Bloomingd~le,
I11lnois. The output of the high llmit compsrator is conrlected to the
cathode of the high limlt dlode 76, and the output o~ the lo~v limlt cornF~ara-
_3L~.

ror i8 connected to the anode of the low llml~ dlode 77. The anode of
the high llrn~ cliode 76 and the cathode of the low limlt diode 77 are con-
nected together and form the ~aturatlon override signal 7~. If the
deslred va1ue 8ignal s~lpplied to t~le hig,l~ llmit c~mparator i9 le~ than
the hl~h limlt ~et point, tl~en tlle higll litrlit cornparator "oes to its hlgh
stflte causin~ the high limit dlode 76 to go to a nonconductive state allow-
lng normal operation.
Similarly, lf the desire~l value 18 greate~ than the low llrnit se~
point, the low limit con-lparator 75 goes to it3 low state and the low lirnit
diode 77 goe~ ~o its noncorl~uctlve state allowing normal operation. If
both circuit~ ~llow normal operation, the error and rate ampllfier circuit
operate~ normally.
.
However, if the desired value is above th~ hi~h limit set point,
the hlgll limit comparator will go to it~ low state cau~ng the high lirllit
dlod~ 76 to become conductive supplyin~, a saturatlon overrlde slgnal 78
to ~he ~rror and rate ~mplifier clrcult Rnd ultimately to the corrective
action circuit to be deA~cribed,
Also, iI tlle desired value i8 less thfln tlle 10~N limlt se~ poitlt~ the
low llmlt comparator will go to it~ low state causiDg tile l~w limit diode 77
to become conductl~e and supply a saturation overrlde ~i~,nal to the error
and rate amplifier clrcult shown ln P`igure 10.
Referrln~, now to Figure lO, for the error and rate amplifier cir-
cuitt it can be seen that the saturation override silrnal 78 ls ~upplled to the
positive input of an instrumentarion ampllîier 8~ which may be such as the
~3S~-

-
~6~77
Model No. A052l, al60 manufactured by Analog Device~, Inc. When the
desired value 13 within the high and low limit set points 72 and 73, the
hlgll liMit diode 76 and the low limit diode 77 are both ln their noncon-
ductive state, resulcing ln no saturation override signal 78 being supplied,
thu~ ef~ectively disconnecting the valid range check circult 69 and allowing
the error and rate amplification circuit 70 to operate in its norn al fashion.
Again, xeferring ~o Figure lO, the desired value algnal, which i8
commollly a s~atic s1~nal, is connected to the pos1tive input of a first
operac lon ampllfler 83, the output of which is connected tO the negative
lnput of the instrumentatioll amplifier 82 with a res1st1ve feedback Rl, con-
nected in paxallel w1th the operational amplifler and prov~dlng a signal to
the nega~ive input thereof, Under static condition3 this prov1des what ls
commonly known in the art a~ a voltage follower circuit whereby the voltage
OUtp~lt of the operational ampllfier 83a i~ equal to ~he lnput thereof, whlch
ln thl~ cace 1~ the deslred value ~ignal,
A second voltage follower clrcuit is similarly provided by con-
necting the feedback s1gnal to the po~i~lve lnput of a second pperational
ampllfler 83b, the output of which i8 connected to the resistance R3 with
the feedback resistance R~ being connected between the output and the
negative ~nput thereof. The resistance R3, whlch 1~ preferably of a rather
low v~lue, allows the saturation override signal 78 to override the normal
operation of the error plus rate amplifier circuit under predetermined
condltion~, as described previously. With both the voltage follower clr-
cuits ~f~ectively connected to the instrumenta~ion amplifier 82, and with
the saturation override signal 78 effectively eliminated a~ described

7~7
above, and wlth the sy~tem ef~ctlvely In a statlc state conditlon, the
correctlon sigllal i9 equ~l ln magnitude to the difference Det~een the
feed~ack and the de~ired value ~Ignal, multiplied by the r~te and pro-
portional gain factor. We, in effect, now llave the 6tat~c state correc-
tion ~lgnal ~hich i8 ~upplied to the corrective action cilcult for the pur-
po~es prevlouELy described.
Howe~er, a dynamic ~tate is encountered when the feedback 6ig-
nal 1~ changirlg in relation to the de~lred value signal, which i~ the case
when the proce~s 1~ ch~nging.
In thls case, we ln effect have a serles circuit from the output of
the flrst operatlonal ampllfler 83a through its feedback resistor Rl through
the capacltor Cl through the feed~ack reslstor R3 to the second operational
ampli~ied 83b output. Depending upon the relationshlp between the desired
value ~Ignal and the feedback slgnal, there will be curren~ flow from the
output of one of the operatlonal amplifier circuit~ through che capacltor Cl
and both feedback resistors Rl and R2 to the output of the other operational
amplifler clrcult causing th~ volcage change rate across th~ capacitor Cl
to be the same as the ra~e of change between the desired value signal and
the feedhack ~gnal.
The voltage developed acro~ Rl as a result of the current flow
will be added alge~raically to the desired value signal voltage and fed to
the negatlve input of the lnstrumentation an~plifler 82. Similarly, the
voltage developed acros~ R~, ~hlch will be of opposite polarity~ will be
algebralcally added to the feedback signal voltage and fed through reslstor
R3 to the posltlve lnput of sald instrumentation a~nplifler,
~`7-

7~
D18-1786-5~
*he instrumentation amplifier ~2 provides as an output
a single correction signal which is a ~unction of the difference
oE the desired value, the feedback signal, the gain Eactors, the
value of the capacitor Cl ancl the rate oE the change between the
desired value signal and the feedback signal. ~his can be
expressed in the formula that the correction signal is a function
of:
G[{F-DV3 ~ Cl x {Rl + ~2} x {d{F-DV}]
dt
where Cl = value of Cl in faracls
G - rate + proportional gain ~actor
F = feedback signal voltage
DV = desired value signal voltage
d = derivatlve of with respect to time in seconds
'-a~E'
R = resis-tance in ohm~
The value of the resistances Rl and R2 will depend upon
the particular process and the desired proportional gain an~ rate
gain. In this particular embodiment oE the error and rate
amplifier circuit, the rate plus porportional gain adjust will be
set Eor the proportional gain desired for the particular process
~ being controlled. Then the variable resistances Rl and ~2 will
`~ be set, preEerably equal to each other, at the value such that
the overall rate gain will be equal to the product of the rate
plus proportional gain factor times the rate gain Eactor.
In this particular mode, which is a difEerential mode,
operating our novel con-troller with the use of relatively high
gain factors, such as the one used by Applicants in one
application of the present invention hav-
- 38 -
~'
~,,,,~,

~ S~777
lng a value of 5, the circuit can e35ily gO to a s~turated conditlon, thus
ma'clng the above formula for ~he correction signal inoperable. Slnce It
1~ deslred to have such formula operable over a~ l~rge a r.~nge as pO3-
~lble, by use of thl~ novel ~rran~emellt of clrcuitry we are able to brlng
tlle clrcui~ out of the satur.lted condltloll ~y use of the rate portion of ~he
clrcult, which ls, in ef~ect, a look Rhead ~eature, nluch earlier than the
proportlonal circuit itself could ~e brought out OI the saturated condition9
~u~ glving much greater controlabillty of the clrcult than wa~ pos~ible
heretofore,
To more fully ulldsr~tand the operation of the error and rate
ampll~ier clrcuit, we should analyæ the correction slgnal output functlon
as defined in tlle formula above, It should ~lsc) he un~erstood ~hat typical
oper~tlonal amplifler~, such as those ~ho~hn as 83a and 83b in Figure lO,
and a typical in~rurnentation ampllfier, such ~ that ~hown as 82, al80
in Flg~re 10~ reaeh the~r s~turated state at approximately 2 volt6 less ~han
the power ~upply voltage furni3hed ~lem. In a typical ca~e, ~he saturflted
~eate occurs at approxlmately ~13 volte DC. ThlY 1B to mean, any input
greate~ than l3 YoltQ or less than '13 vol~ may no~ entirely J~e useable and
no output wlll exceed 13 volts nor be less than -13 volt8. The typlcal feed-
back signal ~roltage a;ld deslred value slgnal v~ltage nre in the range of
æro to S volts DC::, althou~h o~her voltages and other opera~:lonal ampli-
fiers and inatrumentatlon ampliflers flre ~vailable that th~ould result ln
other useable vo1t~ge range~.
~ eferxing to the above ~ormul~, ln a statlc conditlon, the value
of d (F-DV) equal~ æro slnce there i8 no change with respect to time ln
dt
,~9.

'~ >77
the feedback and deslred vaIue signal8. A8 such, the correct~on s~gnal
becomes a functl~n ~f
G x ~ DV)]
when the galn factor, for example. has a value of 10, and wllen the ~if-
ference between ~lle f~ed~cl; ~nd de9ired value signals ~xceeds approxl-
mately I. 3 volts. lnstrumentation amplifler 82 hecomes saturated and the
effect of the correction slgnal i~ to cause ~he process devLce to move to
an extreme condit~on at a ra~id rate, preferably one that ~he process
correlate slgnal can continuously respond to.
When in a stable and static condition there will be
no saturation override signal 78 and the difference error
between the feedback signal from the feedback signal device
42 which relates to the process correlate signal and the
desired value from the desired setting device 41 is less than
the preselected de3dband there is no movement of the process
device 4~. If the desired value is within the set points 72
and 73, the error and rate amplifier circuit 70 will operate
normally, resulting in the appropriate correction signal
be~ng supplied to the corrective action circuit 68 to operate
the driver 43. HoweverJ if the desired value is outside the
valid range set points, this will cause the error and rate
amplification circuit to become saturated and go to a full
plus or full minus saturated condition depending on whether
the desired value was outside the high limit set point 72 or
the low limit set point 73. This7 in turn, will ultimately
cause the process device 46 to rapidly go to one extreme or
another, for example, fully opened or fully closed, and stay
there until some further signals are received from the
circui~ry.

'777
In the typical operatlon, the process controller utll~zes the feed- -
back and desired value slgn~ls whlch are initlally equal In value, for
example zero volts. Thus, the correction signal equals zero. The
desired value slgnal l~ chen suddenly changed to another value wlt hin the
~alld range, such as 3 volt~ DC, whlch causes the correctlon signal to
a~t~ to beco~ne saturated. In this case, since thls ~s momentarily a
s~at~c condltion, the correction signal attempts to become
10 ~ (0-3) = -30 Volt~
However, be~ng beyond ~he saturation llmlt, lt In fact becomes -13 volts
typically~ resultlng in attempting to move ~he process devlce, such as
a carbu~etor throttle, full speed towards the wlde open thrott le po~ltlon.
A8 the process devlce moves, the process correlate s~gnal starts to
increase. We should now reanalyze the above formula by uslng a slightly
different forn~ namely
~ ~ ~(F + G2 d(Ft DV)) (I )V ~ G2 d(F-Dv))]
;~ ~ where G2 = ~1 Cl, and for example m~ght equal 10.

`~ l
777
The factor F = G2 d (~-DV) l~ the output of the second operational
ampllfier 83b, while ~he Iactor DV - G~ d(F-DV) iB the output of the
flrst operational amplifier 83a, nelther of which can exceed the satura-
tlon lirnit, typically 13 volts, Also, the value of the entire formula
cannot exceed the satura~ion llmlt.
As the process correlate signal~ and thus the feedback signal
~ ~tart~ to increase" the value of ~he left portion of the above formula
which is the output of the second operatlonal ampliiier, increases ln
value from zero volts, and the value of the right portion,whlch is the
output of the first operatlonal amplifier, increa~es in value from 3 volts
at a somewha~ slower rate since the value I~V is s~atic. This results ln
an overall reduction in the magnltude of the output of the correction ~ig~
nal from -30 voltq untll the system becomes ~itllln saturation. It should
be observed that ~he maln factor in changing che correction slgnal ls the
factor a2 ~) whlch equ~te~ to the rate of change be~ween the feed-
back and deslred value ~ignala. Thls factor typically might be changing
at ~ spe~ ten times that at which the feedl~ack ~ignal migllt change. A~
such, the correction slgnal is reduced a~ a rate much faster by also using
the x~te of change of the actual error between the feedback and desired
value slgnals then if the error difference only was consldered. This is
termed the look ahead feature, wherein the effecc of the rate of change
between the feedback and de~ired value slgnals is a much laxger factor
in determinlng ~he correctlon ~gnal than the error difference betweerl the
feedback and desired value s~gnal~. When the correction signal falls
hlowex
wlthin the ~aturation vo1tage, the process starts changing flt a
~2 -

-~L4~7~
rate, although the process correlate ~ignal reaponse from the proce~s
18 somewhat slower than the process device because normal opera~ion
of the carburetor,for example, is somewhat ~lug~ish in n~ture.
As the proceQs continues to change at a contInuously slower
rate, the correctlon signal value ehanges to a value wlthin the deadband,
thereby scoppirlg furthex process device change A~ che process cor-
relate slgnal, and thus the feedback signal, con~inues to change some-
what, the correction sIgnal reverses pvlari~y, and a process device
chan~e starts to occur in the opposite dlrectlon, although at a slow rate
BinCe tl~e rnagnitude of the correctlon signal typlcally remains ~mall.
Thls demonstrates a procefis ~evlce overshaot wlth little or no proce~a
overQhoot yielding a faater proce~s acquIsi~ion tlme, chus faster process
eontrol.
In anothe typlcal operation In whlch an external means, ~uch
as throttle adjustment, i8 eausIng a process, such as controlling hood
pressure, to change at a relatively steady rate, the proce3s s~arcs with
the pxoces3 being controlled. Thus, the feedback and de31red v31ue sI~-
nals are ln a static condi~ion and areequal ln value, and thus the cor-
rectlon slgnal equal~ zero, In thls case, the desired value Is held at a
constant value, but the external mean~ of tnroctle adjustmerlt is used to
change the proce3s and ultimately the process correlate signal, and thus
change the feedback signal by for example 0.25 volts per second if no
corrective actlon were to be taken. AgaIn, as this i8 momentarlly a
static condition, the correctlon signal becomes some non-zero value.
This results in moving the procesQ device, such as che hood pressuxe

value, in such a manner as to attempt to keep the ~eedb~ck 6lgnal at it~
deslred value. As the changes of throt~le ad]ustment and hood pres~ure
value e~re occurrll~g, the correctlon s1gnal tal~es on a value SUC}I that the
pl^ocess op~rator ten~s to move at a rel~tlvely constant ~qpeed ln tracking
the feec~back si~nal chanL~e caused l~y the throttle adjustment. Thls cor-
rectlon signal tends to be lndependent of the d(F-DV) function, sincethe
process correlate slgnal 19 esqent1ally maintaining a value somewhat d~f-
ferent than its orlginal value. As essentlally constant value, there i8 no
rate of change in the dlfference between the feedback and deslred value
signals. When further thro~tle adjustment is ceased, the tracking ends
~nd the look ahead feature will tend to dampen the process overshoot as
ln the previous example,
In an addltional type of operatlon ln ~ihich the desired value sig-
nal is changed at some relatlvely steady rate, the operatlon of the error
and rate amplifier circuit is some~;hat slmilar to tha~ OI the prevlous
example. The process device will be moving ln such a manner so as to
attempt to change the feedback slgnal at the same rate ~hat the desired
value slgnal is changing, agaln re~ulting in ~h~ d(F-DV~ functio1l essen-
tlally becoming zero ln value, while the F-DV func~ion takes on some
relatively constant value. When the deslred value ch m~;e stopq, the
tracking ends, and the look ahead feature wlll a~ain tend to dampen the
process overshoo~ ylelding a faster process acquisitlon time" thus faster
process control.
In the case where a saturation override aignal 78 is not effec-
tlvely eliminated, and has been supplled to the error and rate clrcult 70,

~7~7
~his ~lgnal, whlch it~elf i8 a saturated sl,~n~l, cau~es the instrumenta-
tlon ampllfier 82 to be driven and held into po~itive or neg~tlve satura-
tlon. The po1arity of che lnstrllmentation aFnplifier 82 OUtpllt correctlon
slgnal will b~ tlle sa~ne as the polarity of the 3~turatlon overrlde ~lgnal.
Tl~is correction si~nal, ~ above, i8 fed lnto one of the corrective action
circults ~hown in Figure~ 6, 7 and 8.
Re~rring now to ~i2ure 11, the operation OI tlle scaling and
meter protection circuit 71 can be described. In this ca~e, we have, in
effect, two volta~e follower circuit~ with curren~ limitlno re~;istors
hefsre the feedback loop. The flrst of th~6e clrcuit~ i~ formed by ~he
first scaling circuit operationnl amplifler &~c nnd the flrst current
1lmiting resistor 8Sa, and the ~econd of these circuits i~ formed by the
second scaling circuit operational amplifier 83d and a second current
llmiting r~i9tor 85~. A scallng resistor 8~ i~ provlded at the outpu~
of the first current llmitlng reslstor 85a. Thus, when tlle desired value
slgnal enter3 the flrs~ scaling circuit operational ampli~er g3c, and the
feedback signal enters the second scaling circult opera~lonal ampllfier
83d, the ~wo operational amplifler~ together provide a difEerenti~l output
which i~ in the form of voltage, which ha~ 11mited current capacity such
that the meter wlll n~t be overranged. Depending upon tbe particular
meter and scaling res~stor 86 used, the deslred deviation meter output
may be obtalned.
~ eferring now to l~lgure 6, which i8 the preferred embodiment
of the corrective action clrcuit 68, if a D(~ 3teppin~ motor i3 to ~ used
as the operator ~5, the purpose of the correctlve actlon circult 13 ~asically

77
threefold~ First to determine tlle ab~olute v~1ue of the correction slg-
nal, second to indicate to tlle driver to be described hereinafter the origl-
nQl polar~ty of tlle correctlon sigllnl, and tllird to suppIy a clock slgnal to
the drlver. It sllould l)e un~erstoocl that the clock sign~l i9 a ~eries of
pulses wherein tlle frequency varie8.
The nl)solute value circult 87, shown In Flgure 16, con~ists of a
plurallty of operation~l amplifiers connected to variou3 circui~ compon-
ents. A first ah~olute value clrcult operational amplifler 83e having ~
positive and negfltive ~nput is provlded. The poQitlvQ input i9 connected
to annlog cornmon through a resistor h~ving a value of 2/3 R as descrl~ed
hereinaf~er. The ne~ative Input of said opel~ltlonal amplifler 83e ls con-
nected to ~ first surnming junctiotl 88~ The correction signal is supplied
to the susnllllng ~unction 89 through a reslscor having a value of ~, and
al80 tO a second summlng junction 89 through ~ resistor havLD~ a value of
2~. Also lnterposed between the first summing junctlon md the second
summlng junction are two res~storQ in series, both havlng a value of R.
A flrst s~eerlng diode 9518 interposed between said two re3i~tors at
junction po~nt 90 with the cathode of sald first steering diode co~nected
to the output of ~sid first absolute value circuit operation~l arrlplifier 83e.
There is al90 provlded a second steering cliode 9~ having its cathode con-
nected ~o said first sum~1ng junctlon 88 ~nd it9 anode connected to the
output of said first operatlonal ampllfier 83e. A secon~ absolute val-le
circult operational arnplifler 83f has it~ negative lnput cor.nected to sa1d
second summing junctlon 89, and its positive lnput connected to analog
common through a second resis~or havlng a value of 2/3 ~, The output
of ~a1d second operAtional amplifier 83f is also connected co sald second
y~
: .. , . . , ,, .. , ~ . .

~umming ~unctlon 89 through a resistor having a value OI 2R, ~nd pro-
vldes an output signal havlng an absolute Yalue of the Input correction
slgnal. A third absolute value clrcult operational amplifler B3g havln~
its negative Input connected to the output o~ sald first operational
amplifier 83e 18 provlded. The posltive lnput of sald third operatlonal
amplifier 83g is connected to analog common ~hrough a resistor having
a value of ~, and a feedback loop ~9 provided whereln there is inter-
posed a re~istor of value lOR. A polarity signal Is taken off the output
of said thlrd operatlonal ampliIier 83g.
It ls well known In the art that one does not want ~o operate an
o~eratlonal amplifler at lt~ maximum current rating continuously because
its reliabllity 6uffers a serlous drop. Also, one does not want to operace
it at too emall a current because then such factors as nolse, bias cur
rents, and o~her con~lderat~ons come Into play~ We prefer to operate the
operatlonal amplifiers at approximately 10~ of thelr rating, and would
choose Lhe various reslstors in the circuit to so limit the current. In
order to do thi~, the value of any particular resistor would follow the
relationship ~hown whereln the reslstors are ra~ed from ~ to 10R wlth
varlous values in between.
When the corre~tlon signal enters the absolute value circuit 87,
the correction slgnal volta~e 18 applied to the reslstor R assoclated with
the fi~st absolute value circuit operatlonal ampllfler 83a. For a correc~ion
signal voltage greater than zero, che first operational ampllfler clrcuit In
effect has a gain factor of minus one and wlll cause the output of said cireult
at ~unctlon po~nt 90 to becorne the negat~ve value of the Input correctlon
~IgnaL The
~~'
. . .

second operational ampllfler circult associated with summing ~unctlon 89
effectlvely provides an output voltage e~ual to the negatlve sum of the
lnput correction voltage and twice the voltage at junction point 90. In
thls case where the input correction voltage is positlve and the voltage
at ~unctlon point 90 is negatlve, the output voltage 19 -[CV ~t 2(:3~V)~ = tCV
where CV Is a correctlon voltage greater than zero.
However, when the correction signal voltage i~ less than zero,
the voltage at junction point 90 would become tne posltive value o~ the
correction signal voltage except that now the steerlng diodes glve the
first operational amplifier circuit an effective galn factor of zero. This
result~ in the voltage at junction point 90 becoming zero. Now the output
of the second operational ampllfier circult is -[CV + 2(0)] ~ -CV where
CV iB a correction voltage less than zero. Tnerefore, the output of the
second operatlonal amplifier clrcult is a positive signal equal In ampli-
tude to the input correctlon voltage which is commonly termed absolute
Palue.
Slnce the output of ~he fkst operational ampllfier 83e between
the two ~teerlng diodes will always have the opposlte polarity of the
Input correctlon signal, the negatlve polarity signal Is fed to the negatlve
input o~ the third operatlonal ampllfler 83g whlch, in effect, acts as a
comparator. The output of the third o~eratlonal amplifler 83g is caused
to be saturated ln the opposite polarity of lts lnput since tlle reslstors lOR
and ~ were chosen to obtaln said ~aturated condltion. This gives U8 a
potarlty slgnal as Indicated in Flgure 6 with the same polarlty as the
correction ~Ignal.

77'7
The absolute value ~Ignal from the absolute value circuit 87 ls
then ~upplied to the dead band comparator 92 whlch may be such as
model No. AD311 manufac~ured by Analog Device~, Inc. prevlously men-
tloned. The function oî sald dead band comparator i9 to compare the
absolute value of the correctlon slgnal wltll dead band reference values
whlch have been 8upplled ~hereto by any ~uitable means. I~ the absolute
value of the correctlon ~ignal X Is between zero and the dead band refer-
ence ~alue, the dead band comparator act~ to cause the proces~ device
46 to remaln In Its present posit~on by disabling the clock output. How-
ever, if the absolu~e Yalue ls not between zero and the dead band refer-
ence value, the ab801ute value of the correction sLgnal is then supplled
to the ~ummlng ampllfier 91 shown ln Figure 13,
Summing ampl~flers are common In the art and the component~
thereof, or lts operat~on, need not be described herein In detall. It Is
to be noted, however, that the transfer functlon for the pa~ticular circuit
used ln thls summing ampllfler resulL~ In the equation: Output = -Rf (Ra
lRxb 1 ), Thus, we now supply the 6ignal from the summlng amplifier 91 ~o
the voltage ~o frequency c onverter 93 which may be ~uch as the model
No. .4.~537 manufactured by Analo~ Devlces, Inc, of 8100Mingdale, Illinoi~,
or a~y of several other dev~ces known In the art. If the dead band com-
para~or 92 has not prevlously caused the analog switch 94 to disable the
output from said )~ converter 93, a clock ~Ignal will be supplled to the
drl~er 45. The analog swltch may ~e such as the model No. AD75 13 man-
ufactured by ~he aforementioned l~nnlog r)evices. Inc., or could he an
equLvalent transistor clrcult well known in the art,

6~7~
The clock slgnal and the polarlty slgnal belna supplied to the
drlver wlll ultlmately be transferred to the operator 45, whlch in thls
case ls a DC stepping motor, and wlll control the speed and dlrection
at which sald motor operates. Slnce the correctlve ac~lon circuit shown
in Figure 6 ~g partlcularly adapted for drlvlng a l~C stepplng motor, a
stepplng motor driver must ~e u~ed in conjunction therewlth. There are
many stepping motor drlvers such as those manufactured by the Superior
Elect~lc Co. of Brlstol, Connectlcut and Slgma Instruments, Inc. of
Braintree, Massachusett6. However, the preferred-embodlment of the
present Inv~ntlon when a DC stepping motor Is to be used, conslsts of
a steppe~ transla~or connected to a quad 5ADC driver. These unlts are
avallable commerclally from Scans Associates, Inc., oE Livonla, Michlgan,
as stepper translator model No, 30086 and quad 5~DC driver model
No. 30083. We have found this partlcular drlver system to be very ad-
vantageous because of the fact that lt 1~ a higher performance system
than others commercially av~lable, and it has several other feature~
sucll as full or half ~top operatlon, polarlty reversal, and optically iso-
lated outputs and inputs, whlch are very desirable ln reducing nolse effects
In the sy~tem and allow~ng interconnection with and around machine con-
trol apparatus. Also, if desired, in place of the valid range check clr-
cuit 69, llmit switches could be connected to this preferred driver system
to prevent the ultlmate process operating device 46 from exceedlng the
fully opened or fully closed type posltion~
If for rea~ons such as spee~, torque, cost of the particular
applicatlon or the like, the drivers so far descrlbed, whlch are all DC
. ~, ... ._., .. _ .... .

17S~6-5~
in nature, may not ~e applicable, it may be desirable to use a
standard reversible motor other than a DC stepping motor in an
incremental or step mode. Such a motor would normally be an AC
motor i~hich would require, in additlon to the corrective action
circuit shown in Figure o, in turn, a two directional switched
driver which is shown in ~igure 17. In this instance, a divide
by N circuit 103 is provided which may be the same as a Motorola
model No. MC 14522B or its equivalent. This circui-t has the
clock signal connected to one input, and an N assignment device
104, which may be a thumbwheel switch or other suitable switching
device, connected to the present inputs. The output oE the divide
by N circuit is connected to a retriggerable timer 105 which may
be simllar to Motorola model No. MC 14528B or some similar
device. This particular timer has proven to be desirable because
it is of a programmable nature having provisions for an increment
duration or magnitude adjustment. The output oE the timer 105 is
connected to one input each oE a first -two input and gate 111 and
a second two input and gate 112. The polarity signal from the
corrective action circuit is connected to the second input oE the
second two input and gate 112 and is also connected through an
inverter 110 which may be such as Motorola model No. MC 140~9B to
the second input of the first two input and gate 111 in the
manner shown in Figure 17. The output oE the firs-t two in~ut and
gate 111 is connected to the base of the Eirst driver transistor
113. The emitter of said first driver transistor is connected to
the logic common and the collector thereof is connected to a
~irst driver relay 115 which may be such as the model No.
6563~-22 manufactured by ~athaway Controls of Tulsa, ~klahoma.
The contact connections from the first driver ~elay may be used
- 51 -

777
in many ways, three of whlch wlll be descrlbed below ~n r~gard to Flg-
ure 18 through 21,
,
Slmilarly, the outpu~ of the second two lnput and gate 112 18
connected to the base of the second drlver transistor 114 which may be
ldentlcal to the fir~ driver transistor as ls the case in the present embodi-
ment. The emltter thereof l~ agaln connected to loglc common with the
collector being connected to the input of a second drlver relay 116 whlch
may be identical to the first, if deslred. The contacts îrom the second
driver relay 116 can be also used for any desired purpose. One particular
use of the contacts from the flrst drlver relay and the second driver~relay
whlch we have actually used 18 to connect them ln the manner shown In
Flgure 18 ~o an AC synchronous motor such as the model No. SS4OORC
manufactured by Superior Electrlc Co. of Brlstol, Connectlcut.
It should be understood, and will be understood by one skilled ln
the art that many of the components shown ln the figures for which model
nurnbers have been supplied can be substituted by many other substantlally
identical corr4~onents having other model numbers and belng manufactured
by other ~nanufacturers, and the cLrcultry of the presen~ inventlon wlll
perform as desired. Only the preferred embodiment has ~een shown
hereln, and sorne of the reasons for such preference have been given.
l~her reasons havlng to d~with avaUabillty, cost, size, etc. alsowere
taken lnto account by the Appllcants.
It ls contemplated that when a su~stltution Is made, after appro-
prlate subs~ltution guldes have beRn consulted, wirlng dlagrams fo~ ~he
-5~-
., , ,, 1 1 .

777
particular devlce belng gubstltuted may be ea3~1y obtalned from the litera-
ture supplled by the manufacturer of the partlcular devlce belng used.
Also, ~t should be understood In regard to Flgure 18 that the
contacts from the ~irst and second drlver relay can be used in many other
ways other than connectlng them to the particular AC motor wlth which
Applicants have experlence. Examples of such uses are the use of mo~t
any reversible motor, or two dlrection actuator to control mechanlcal,
pneumat~c or hydraulic clrcults. Such actuator m~y be rotational or non-
rotational In nature.
Referrlng agaln to Figure 17, our ~wo directlon swltched drlver
would accept the input of ~he clock and polarlty slgnals and the N lnput
~upplled by the N asslgnnlent device 104. The dlvide by N clrcuit puts
out one pulse for eversr N input ~ulses and this serves to scale down the
high frequency clock rate pro~lucing the lncrernent rate. The 6caled pulse
rate l8 then u8ed to trlgger the retrlggerable tlmer 105. The tlmer out-
put is then gated wlth the a~ove-mentioned polarlty ~Ignal to produce
separate forward an~ reverse output sLgnals by means of the firs~ and
second two lnput and gates, the flrst and second driver ~ransis~ors and
~he first and sec~nd driver relay~L The slgnals, v~hich are In the form
of contact closure~ as previously mentloned, may be used to drlve most
any motor or two directlon actuQtor by way of standard swltcllln~ techniques.
The increment mEIgnltude adjustment i9 used to determine the duration of
contact closure for each N clock pulses.
~.... . ........ . . . . . . . . . .

7'7
A use of our two dlrection swltched drlver for controlllng a
DC motor may be such as that shown in Figu~e 19 wherein the relay
con~act 115a whlch ls understood to be the contact of the flrst driver
relay 115 and tlle relny contact 116a, whlch Is understood to be the
relay contact of the second drlver relay 116, are connected In the
manner shown to a standard l~C motor.
If Lt 19 desLred to operate pneurnatic or hydraullc cIrcuits
incrementally wIth out two dLrectlon ~witched drLver, the method of
use illustrated ln Figures 20 and 21 have been shown to be satisfactory,
whereln the fLrst drlver relay contact 115a and the second drLver relay
contact 116a are connected as shown in Figure 20 to a solenoLd A and a
`~ solenoid B of a double solenold value which are, Ln turn, connecced to a
pressure operated cylinder 118 in the manner shown In Flgure 21. When
solenold B is operatLng the position of the dou~le solenold valve shown Ln
Figure 21 causes pressure co enter the left-hand end of the cylinder 118,
causlng the piston thereo~ to move to the right and the cylinder to extend.
When the solenoLd A Ls operatlng, the valve shif~s positlon causlng the
plston to move to the left and the cylLnder to retract.
However, in certain processes it ls deslrable to use pneuma~ic
control actuators such as the operator 45. ThLs requires some changes
in the correc~ive actlon circult and results ~n the embodlment shown Ln
Figures 7 and 8, When the pneumatIc corrective action circult shown in
Flgure 7 Ls used, the correctlon ~l~nal from the differential Lnput t~ircult
67 fLrst passes in~o an absolute value cIrcuit 87, which Ls {dentical to
that previously described in Figure 16. The ou~put of the absolute value
...... .. ~ .

i7~7
clrcuit agaln is the absolute value of the corrective actlon signal and
th~s i8 pa~sed lnto the dead band comparator 92. The polarity output
from the absolute value clrcuit ls not used in this embodiment. In a
manner slmllar to that previously descrlbed, the absolute value of the
correction ~ignal wlll be compared wlth the dead band reference and lf
it 18 between zero and the dead band re~erence the analog swLtch 94 1~
d~abled. Therefore, no current can flow into the integrator 98 and no
change in the output of the pneumatlc corrective action circult OCCU~8,
and tbus th~ slgnal to the drlver 43 ls effectiveay froæen.
However, if the absolute value of the correctlon signal is greater
than the dead band reference, ~he analog switch 9~ lg enabled allowlng
current to flow t~ $he integrator 98. In this conditlon, the correction
signal is ~upplied to the scaling c~rcuit wlllch, in effect, i8 a simple
potentlometer well known ln the art,, Thus, the correction slgnal is re-
duced In value in a predeter~nlned proportion and provide~ a properly
scaled signal to the lntegrator 98.
Referring to Figure 14, ~he Lnput ~o the integrato~ 98 passes
through a reslstor R~ lnto ~he negatlve lnput of the integrator clrcuit
o~erational arnpli~ler 83h. A eedback loop containing a capaci~or CI
~s provided from the outpu~ of the operational ampllfier back to its ne~a-
tlve input wlth its positive input connected to analog common. The ef~ec~
of this is to change the input slgnal into a voltage signal representing the
rate of change of the voltage. The values of RI and CI are chosen to pro-
vide a time constant for $he circul~ such that the process device 45 is
.... . .. ..

77
capable of followln~ the OUtpUt sl~nal tllrough ~he drlver 43., In general,
the output 1~ a functlon o~ l~E Cl ~ and tlme.
The volt~e ~ nal out of the lntegrator ~81~ then passed through
a l~uffer-scaaer 100 shown ln rnore detall In Flgure 12. The buffer~caler
~s, In effecc, a blpolar drlver ~ollo~ver composed of a NPN transistor Ql
~uch as a ~N4921 and a PNP translstor Q2 such as a model 2N4918 wlth
~h~lr b~s~s ~ h ~r.~r.~ct~ t~ t~ Input slgnal supplled from the Inte~ra^
tor 98 and the emltters both connected to a scallng reslstance Rs whlch
provldes an output ~Ignal to the drlver. The collector of Ql ls connected
to plu~ ~CC (power supply voltage) and the collector of Q2 1~ conJlected
to minu~ VCC. TllU8, a ~Ignal 18 provlded to the drlver 43 whlch in thla
case ls aicurrent to pres~ure converter such as a Moor~ Products ~lodel
No. 77 manuf~ctured ~n Sprlnghouse~ Pennsyl~ranla.
In a process where a pneumaclc control ~evlce ~5 and thus a pneu-
matlc drlver i8 necessary and ~ ra~e actlon i5 deslrable, the embodlment
shown In Flgure 8 1~88 proven deslrable~ ln tlli~ case, slmll,qr to tllat des-
crlbed ln connection wi~ Flgure 7j the çorrectlol~ signal from the dlffer-
elltlal lnput clrcuit i8 supplied to the absolute value circulL wl~lch, In the
manner prevlously descrlbed in col~n~ction with ~ ure 16, supplles ~n
output equal to the absolute value of the correction 6i~nal and a polarity
slgnal. The absolute value ~l~nal from the Absolute value clrcult ~9 a~aln
supplEed to a dead ban~ comparator 92, ~nd ~f the al~solute value of the
correctlon slgnal 1~ less ~than a dead band reference, ~he dual analo~
swltch 97, whlch al~o may be such as model No. A~7513 manu~actured
~,6-
.. .. . . . . . . .. . . . . . .

;'777
by the aforementloned Analog Devices, Inc., disa}~les both iaputs to the
sumrning integrator 102, thus resulting in the slgnal to the buffer-scaler
100 belng held constant, whlch ultimately results in no change being
supplled to the operatlng dev~ce 45.
Howevec, if th~ absolu~e value of the correction slgnal i8 greater
than the dead band reference, the analog switch will not dlsable the inputs
~o tbe summing lnte~rator 102. In thi9 case, referrlng agaln to Figure 8,
the correctlon 6ignal is slmultaneously fed to the scaling device 99, which
may be identlcal to that shown ln Figure 7, and 1s, in effect, a potentiometer~
This results in some change irl magnitude of the correct1On signal being
supplled to the analog switch. The saturated polarity signal from the abso-
lute value clrcult 87 1B simultaneously belng supplled to a second scaling
derice 101, resulting In a second lnput to the analog swltch 97. This
second slgnal will basically ~e a constant posltive or negatlve signal depend-
ing on the polarlty ~Ignal. With the analog switch In ~ts enabled condltlon,
both of these inputs are supplied to the summing integrator 102 such as -
that sbown in Figure lS, The summing Integra~or consists of a sumrning
integrator clrcult operational amplifler 83i having Its posl~ive input con-
nected ~o analog common and a feedback loop baving a capacitance Csi
interposed between its output and its negative inpu~. The two input slgnals
from the scallng devices 99 and 101 pass through the resistances Rsiland
RSi2~ respectively, and are connected to the negative input. The values
d the reeistors and capacitors are again chosen in view o~ the consldera-
tions prev1Ously dlscussed deallng wi~h the integrator shown in Figure 14
and depending upon the partlcular appllcat~on to whlch the process
3--~

'777
c~ntrollcrSst~bepu~. The ou~utofthe sum m1ng1ntegrator lO~1~ a
funct1On of ~1 ~ Y2 and ~im~ Thl~ volta3e 81gnal
1~ . 812 61 -
~is supplied to the buffer-scaler lO0, which performs the same
operation on the signal as described in relation to Figure 7.
It can be seen that Figure 8 is substantially similiar to
Figure 7 excep~ for the second scaling device lOl. The func^
tion of said second scaling device is to provide a voltage
input that effectively gives a minimum speed signal to the
driver 43, causing the process device 45 to move at minimum
speed thereby creating a reset type of action when outside of
the deadband range. In a manner similar to that previously
described, the driver may be such as a Moore Products current
to pneumatic converter model 77. The driver, in turn, sup-
plies a signal 48 to the process 44 as shown in any one of
Figures l to 3, and the process correlate signal is continu-
ously compared to the desired value signal until the process
is within the desired limi~s, thus completing the loop for any
of the devices described ~thus providing a novel single-state
four-mode controller which controls a process as a function of
the difference of, and rate of change between, a desired value
and a current state of a process.
Now referring to Figure 22, there is shown a typical
use of our improved three-state four-mode process controller
generally designated by the numeral 125. Similar to that pre-
viously described with Figure l, the process controller is
supplied with a voltage reference indicating a desired value
from a desired value setting device 160, which causes the pro-
cess controller to supply a signal to the driver 43 which, in
turn, supplies a process input signal 48 to the process gen-
erally designated by the numeral 44. Since this is a closed-
- ~d~ -

77 7
loop system we are concerned with, the process 44 will then
supply a process correlate signal 49 indicating the current
state of the process. If the process correlate signal is a
voltage signal useable by the three-state process controller
125, it may be directly supplied thereto. If however the
process correlate s~gnal is not directly useable, a feedback
signal device 42 is needed to convert the signal into one
useable by the controller. For example, if the process cor-
relate signal 49 is pneumatic in nature, the feedback signal
device may take the form a pressure transducer.
As mentioned previously7 such means for converting the
signals are well known and no additional description of the
feedbacX signal device 42 is deemed necessary herein.
Since we are now mainly concerned with controlling a
wide variety of processes all of which might necessitate set-
ting the process at many different desired values, Figure 23
shows an embodi~ent of our invention where it is desired to
automatically operate at said variety of desired settings,
such as to move a control valve over many test points in a
system which is designed to control the manifold vacuum in a
carburetor testing system such as shown in Figure 27. In such
case for a typical carburetor test one may test at as many 20
or 30 points. Some modification is preferred for this situa-
tion over the generalized version because you would need a new
desired value from the desired value setting device 160 for 2
each test point. While these could be set manually, as
~,~

i7~7
will be discussed below in relation to Figure 24, it is much
easier to have some sort of automation device 184 which will
automatically change the desired value for the next condition
similar to that described in connection with Figure 2. It is
also possible, as shnwn by the dotted line in Figure 23, to
tie the cutput from the feedhaçk signal device 42, or the pro-
cess correlate signal 49, to the automation device 184 as
before. This may be desired to confirm that the particular
condition at which the process has arrived is indeed the de-
sired condition before the automation device 184 takes further
action.
As shown in Figure 24, the manual system is in many
respects similar to the system shown in Figure 22 and 23 except
the automation device 184 is elimina~ed and the desired setting
device 160 is replaced by a potentiometer 55, which is used in
the manner previously described, and by a pushbutton switch
161 which is used to reset th~ three-state process controller
to its first state as will be described herein.
An improvement in a system which could be used either
with the three state four-mode process con~roller being
described or the single-state four-mode con~roller previously
described or indeed with any of the systems previously de-
scribed wherein the controlling of the hood pressure, for
example~ is concerned is shown in ~igure 25. In this case,
and in addition to the driver 43, operator 45 and process
- ~0 -

device 46 ~here is a second driver 126 whose input is con-
nected to the three-state process controller 145 and whose
output is connected to the input of a second operator 127 at
the second process input signal 129. In ~urn the process
speed improvement de~ice 128 has i~s input connected to the
output o~ the second opera~or 1270
In this case, when one is initially employing a pro-
cess 44 in which one is attempting to control the hood pressure
one may have a system as shown in Figure 26. In order to con-
trol the hood pressure inside the hood S9 one must first
measure the hood pressure, and this is done by an absolute
pressure transducer 47b which has previously been identified
as the 1332 series manufactured by Rosemont Engineering of
Minneapolis, Minnesota in reference to Figure 4b. In a manner
well known in the art, said absolute pressure transducer pro-
duces the process correla~e signal 49 which in a manner
similar to that previously described is fed through the
feedback signal device 42, if necessary, and then fed into the
three-state process controller 125.
As previously described, the process correlate signal
49 would be compared with the feedback signal, in a manner
shown in Figures 22 through 25, with a signal from the desired
setting device 160, and if a difference exists between the
actual status of the process and the desired status of the
process, the process input signal 48 from the driver 43 would
be used to drive the operator 45, which in this case is a
~, /

,~ 4~7~r7
valve operator 45b, ~hich drives the process deYice, which is
in the form of a valve 46b, to a new position. ~owever, a
second signal would be supplied to the second driver 126 which
in turn would supply a second process input signal 129 to a
second operator ]27 which in this case is in the form of valve
operator 127 driving the process speed improvement device 128
which is usually in the form of a valve. This would be done
any time the` desired hood pressure is much lower than the
actual hood pressure because of the relatively large air vol-
umn under the hood. The throttle plate 152 of the carburetor
56 in most cases will be in a position which substantially
restricts the carburetor throat 151 and thus an extremely long
time will be needed for the vacuum supply to pull sufficient
air from under the hood 59 to reduce the hood pressure to the
desired value. In the preferred embodiment ~he process speed
improvement device 128 sho~n as a valve in Figure 26 would
snap completely open whenever a reduction of the hood pressure
under the hood S9 ~as called for and would stay completely
open un~il the new desired hood pressure is reached, at which
time the valve 128 would snap completely shu~ after which time
the three-state four-mode process controllerlwould operate to
make the final adjustments to obtain the desired hood pressure.
This would again be done by continuously supplying ~he new
process ~orrelate signal 49 to the three-state four-mode pro-
cess controller 125 through the feedback signal device 42, if
necessary, then comparing said feedback signal with the desired
Yalue signal from the desired setting device 41 and~ if neces-
sary, supplying a changed signal to the driver 43 which would
~.~

3L~ ~'7~7
again produce a new process input signal 48, with the opera-
tion continual].y repeating itself until the desired value is
reached within tlle selected deadband limits. The actual con-
nection o the second driver 126 and second operator 1~7 to
the process speed imFrovement device 128 within the process 44
are well l.cnown ir the art and need not be described further
herein.
A basic system which may be used embodying our three-
state four-mode process controller is shown in Figure 27. The
basic systems shown in Pigures 27 through 31 are for testing
carburetors in a laboratory environment wherein the control of
hood pressure, manifold vacuum and air flow is required. In
operation the carburetor 56 would be mounted under the hood 59
to ~he riser 57 in a manner previously described. The hood 59
is shown in its closed position but of course it should be
understood that the hood 59 would either be manually removable
from a suitable test stand or an automa~ic means of opening it
would be provided. Needless ~o say the space under the hood
5~ would be sealingly enclosed so that outside conditions
would not influence the carburetor test. The next step in a
carburetor tes~ utiIizing the present invention is for the
manifold vacuum measurement and control system 135 to cause
air to flow from the air supply (not shown) through the hood
pressure and control system generally designated also by the
numera~ 135 as they may be identical sys~ems from a physical
construction point of view as will be discussed below. The J
- ~3-

~ ~ 4 ~77
air will then flow through the air flow measurement and con-
trol system also designated by the numeral 135 for the above
stated reason. The air will then flow through the conduit
137 to the space enclosed under the hood 59, through the car-
buretor throat 1~1 and in ~urn through the conduit 136 to the
manifold vacuum measurement and control system 135 which is
connected to a vacuum supply (not shown). Air flowing through
the carburetor 56 draws fuel into the carburetor through the
fuel line conduit 153 which is connected to a fuel flow
measurement system which may be such as is readily available
in ~he art.
It is not fel~ that the vacuum supply need be described
in detail, as the vacuum source is normally in the form of a
vacuum pump of which there are many types on ~he market. It
should be understood tha~ any vacuum pump may be used pro-
viding it is of sufficient size to produce the air flow
necessary through the carburetor being tested so that all
desired tests can be run. In this regard it should be noted
that it is necessary to consider whe~her there are sonic noz-
zles to be run or ~he system is to used in a subsonic condi~ion
in selecting the vacuum supply sys~em.
Similarly, the air supply need only be a source of air
which is being controlled as to temperature, pressure and
humidity. Many air supply systems are available and again any
o~ such systems may be used provided they have a sufficient
- 6~ -

~7
capacity to flow the desired amount of air through the carbur-
etor being tested so that such carburetor may be tested under
all desired conditions. Also an adequate fuel supply system
must be used in conjunction with the fuel flow measurement
system.
To proceed with the details of the carburetor test,
the manifold vacuum measurement and control system 135 has
caused air to Elow through the carburetor 156. Depending upon
the test specifications, the hood pressure measurement and
control system 135 will usually keep the pressure under the
hood 59 at a pressure near sea level or at a pressure equiva-
lent to a certain relatively high altitude such as that at
Pikes PeaX. In performing a carburetor test in the laborator
one must set the desired values of hood pressure, manifold
vacuum, and air flow for each flow point at which it is desired
to test a carburetor. In order ~o obtain ~he fastes~ ~est
speed and proper test condi~ions 9 it is desirable that the air
flow, hood pressure, and manifold vacuum control and measure-
ment sys~ems operate simultaneously without causing hunting or
oscillating type of control in any of the systems. It should
be recognized that the air flow measurement and con~rol system
135 will cause the throttle plate in the carburetor to be con-
trolled by the throttle operator 45 and be rotated until the
desired air flow is preset through the carburetor. At this
point then you have achieved a given air flow at a prede~er-
mined manifold vacuum and hood pressure. Having achieved the
desired air flow through the carburetor, one is in a position

~L'~ ~6~
to know the mass air flow rate through the carburetor and if
one now also measures the mass ~uel flow rate entering the
carburetor, the air/fuel ratio o~ the particular carburetor at
the predetermined t~st point conditions can be determined.
In Figure 28, when it is desired to use a hood pres-
sure process speed improvement device similar to that described
in Figure 26~ the condui~ 154 is connected in any suitable man-
ner to the sealed space under the hood 59 at one of its ends
and at its other end to the hood pressure measurement and con-
trol system, which in this case is indicated by the numeral
138 to show that it is no longer identical to the manifold
vacuum measurement and control system. It should be understood
at this point that a process speed improvement device could be
used in many systems where there may be an excessive time delay
usually caused by a large volume of a compressible fluid.
.
This system would opera~e in the manner just described
for Figure 27 bu~ incorporates in addition to the conduit 154,
the process speed improvement device in the form of a valve
128a and the second operator 127 (See Pigure 26).
In this case we are describing a system w~ich is one
embodiment of a carburetor test system utilizing our inven-
tion, and only the air flow and manifold vacuum measurement
and control systems are iden~ical and use our three-state
four-mode controller 125. The hood pressure measurement and
con~rol system 138 also uses a three-state four-mode process
controller, however it contains the process speed improvment
device.
- 6f -

A modification of our invention is shown in Figure 29
which is similar to Pigure 27 but employs a computer system
139 to aid in the test by monitoring the three controller
systems and providing the desired value settings by acting as
the automation device 154. ~ further modification of our
-nvention is shown in Figure 3~ which is similar to Figure 28
but employs the computer system 139 as previously described.
Another embodiment of our invention is shown in Figure
31, which is sim;lar to Figure 28 but employs khe computer
system 139, and utilizes said compu~er system to control the
air flow. In this embodiment it should be noted that the air
flow measurement system is now designated by the numeral 140
as it no longer controls the throttle operator 45, this func-
tion now being controlled by ~he computer system 139. How-
ever, this is only true with regard ~o the air flow measure-
ment and control system, because the hood pressure and
manifold vacuum measurement and control systems are now
identical and both use our three-s~ate four-mode control in a
manner to be more fully described. In this case the computer
system acts as a watchdog type system supplying desired value
signals to the two process control systems, and as an air flsw
control system in response to process correlate signals re-
ceived from the air flow measurement system 140. In mos~
other respects it is similar to the operation described for
~he system of ~igure 28. In addition, ~he conduit 153 is
again sealingly connected to the enclosed space under the
~ j7

hood 59 at one end thereof, and ~o the hood pressure measure-
ment and control system 138 at the other end thereof. Again
~he process speed improvement device would operate similarly
to the manner described in connection with the description of
the Figures 25 and 26 and would require the second driver 126,
the second o~erator 127 and the process speed improvement
device 128.
The descriptions of ~he uses of ~he three-state four-
mode controller thus far described for use in a carburetor
testing system have been described in general terms showing
various closed-loop processes. It should be understood that
such three-state four-mode controllers can be used in vir-
tually any process where a standard process controller, such
as the single-state controller previously described or a com-
mercially available controller can be used. This is true
whether one is concerned with electrical, pneumatic or hydrau-
lic processes, as the method of control would be the same for
all three types of process, only apparatus would be different.
. ' ' .
-~ To more fully understand ~he detailed operation of
our three-state four-mode controller it is to be noted that
the process controller 125 shown in Figures 22, 23, 24 and
25, consists of two portions, the three-state differential
input circuit 145 and the corrective action circuit 68. In
general the three-state four-mode process controller compares
the feedback signal with the desired value signal from the
desired setting device, finds the actual error difference

between the two signals (static), finds the rate of change
(dynamic~ between the two signals, sums them algebraically,
and then provides an output signal related to the error, the
rate of chan$e, a deadband range, and the "state" of the con-
troller to operate the driver 43 as necessary. When in a
stable and static condition there will be no saturation oveI-
ride signal 78 and the difference error between the feedback
signal from the feedback signal device 42 which relates to
the process correlate signal and the desired value from the
desired setting device 160 is less than the preselected dead-
band there is no movement of the process device 46.
For each new set point, the desired value setting
device 160 will now supply a new desired value signal to ~he
three-sta~e four-mode controller 125 as shown in Figure 22.
As before this signal will be supplied to the three-state
differential input circuit 145 as illustrated in Flgure 32
and more particularly to the three-state error and rate
amplifier 146 whose operation will be described later. This
signal is also supplied ~o the valid range check circuit 69
which operates in the same manner as previously described in
~ ., .
connection with our single-sta~e four^mode process controller.
Also this signal is supplied to the scaling and meter protec-
tion circuit shown in Figure 11 which again ac~s in the manner
previously described. For a new set point, the desired value
setting device 160 may also supply a reset state signal to the
~hree-sta~e four-mode controller as shown in Figure 22, in
particular to ~he ~hree-state differential input circuit 145
as illus~rated in Figure 32.
~9

We refer now to Figure 35 which shows the detail of
the three-stat~ error and ra~e amplifier circuit. We have
already described how the satura~ion, override, feedback,
desired value and reset state signals are provided. As in
the single-sta~e error and rate ampli~ier circuit 67, in this
case the desired value signal goes to the positive input of
the first operational amplifier 83a, the feedback signal goes
to the positive input of the second operational amplifier
83b, and the saturation override signal goes to the negative
input of the instrumentation amplifier 82. In this embodiment
the reset state signal is now supplied to the reset input of
the state counter device 156 and the polari~y signal from the
absolute value circuit 87 shown in Figure 16, which operates
in the manner previously described, is supplied to the input
of the edge detector 157. The edge detec~or consists of an
"exclusive-or" gate 158, having a first and a second input.
Interposed be~ween the input of the edge detec~or and the
first input o the "exclusive-or" gate 158 is the firs~ edge
de~ec~or resistor R5.
Interposed between the second input of the
"exclusive-or" gate 158 and the input of the edge detector
157 is a second edge detector resistor R6 also interposed
between ground and the second input of the "exclusive-or"
ga~e 158 is the edge detector capaci~or C~.
It should be noted that t~e polarity signal will go
directly to the first input of the '7exclusive-or" gate, but

will be delayed in ~etting to ~he second input of the
"exclusive-or" gate because of the manner in which the edge
detector capacitor C2 and the second edge detector resistor
R6 are connected.
A pulse output is provided from the edge detector 158
e~ery time the polarity signal at its input changes polarity.
Such output is connected to the clock input of the state
counter device 156 which may be a Motorola, Inc. Model
- MC14017B. Each time a pulse is supplied to the clock input,
the state counter will incrementally advance from the state
it was previously in. Since we use a state counter 156
having a state one output, a state two output and a state
three output, each time a pulse is received the state counter
will provide an output which will advance from state one to
state two or from state two to state ~hree. The reset state
signal is used to rese~ the state counter to sta~e one.
- The reset state signal will cause the state counter
device 156 to initially have a state one output and the
absence of a reset state signal will allow the state counter
to proceed to state two and further to state three. I~ is
necessary to keep the state counter 156 in state three during
further changes to the polarity signal. This function is
performed by the clock inhibit input of the state counter 156
which is connec~ed to the state three output of the state
coun~er thereby latching the state counter into s~ate three
where it remains until another reset state signal is received
at the reset state input of the state counter 156.

:
~ f~ 7
To actually cause the changes in direction of the
process device 46 from state one to state two, and from state
two to state three, the correction signal must have different
values for each state. It should be recognized that while in
state three, the cor~ection signal changes are the same as
described for the cperation of the single-state four-mode
controller.
,~ .
It is now that the use of the state one, state two
and state three outpu~s of the state counter device 156 are
utilized to accomplish this, as they act to connect three
different sets of variable resistances (one set for each
state) between the outputs and negative inputs of the first
operational amplifier 83a, and the second operational ampli-
fier 83b as well as across the gain set inputs of the instru-
mentation amplifier 82.
,
As can be seen from Figure 35, each set of resistances
consis~cs of three separate variable resistors which may be set
to the same or different resistance values as needed to cause
the proper operation of the three states to occur.
It can be seen ~hat when the state counter device is
in state one, corresponding to state one on the graph in
Figure 33, the first state one variable resistor RlA is con-
nected from the output of the first operational amplifier 83a
to the negative input thereof through first state one analog
switch 94c, the second sta~e one variable resistor R2A is
- 7~ -

~s~
similarly connected through the second state one analog
switch 94f across the second operat.ional amplifier 83b, and
the third state one variable resistor R4A is connected across
the gain set inputs of the instrumentation amplifier 82
through the third state one analog switch 94i.
When the state counter device is in state two, first,
second and third state two analog switches 94b, ~4e, and 94h
respectively are brought into action and respectively connect
the firs~ state two variable resistor RlB from the output of
.
the first operational amplifier 83a to the negative input
thereof, the second state two variable resistor R2B from the
output of the second operational amplifier 83b to the negative
input thereof, and third state two variable resistor R4B
across the gain set inputs of the instrumentation amplifier
82, thus forming gain factors for these three devices which
may be different from thsse in state one.
Similarly in state ~hr~e, first, second and third
state three analog switches 94a9 94d, and 94g respectively
are used to respectively connect first state three variable
resistor RlC from ~he output to the nega~ive input of the
first operational amplifier 83a, second s~a~e three variable
resistor R2C from the output ts the negative input of the
second operational amplifier 83b9 and third state three var-
iable resistor R4C across the gain set inputs of the ins~ru-
mentation amplifier 82, fsrming gain factors for these three
devices which may be different from those in state one or
state two.
- 7~ -

m
To correlate these resistors, and to show how the
device goes from one state to another, it should be under-
stood that ~he resistors Rl, R2 and R4 utilized in the three-
state error and ra~e amplifier correspond exactly to the
resistors Rl, R2 and R4 shown in the error and rate amplifier
circuit of Figure 10 for the single-state controller. The
resistor R3 is unchanged for the two different controllers.
It can be seen then that the state counter device 156 in con- -
nection with the edge detector device 157 causes the three-
sta~e process controller to change states as shown in Figure
33. The state counter 15~ is reset to the state one via the
reset signal, and then incremented to state two and to state
three, via the polarity signal and the edge de~ector, where
it will remain until the reset signal is again provided.
The values of the three sets of resistors across the
amplifiers are chosen such that if the state counter is reset
to state one the process device 46 will operate at a prede- -
termined rapid speed in the desired direction. When the
polarity signal changes polarity, the state counter 156 will
receive a pulse from the edge detector 157 causing the state
counter and thus the three-state four-mode process controller
to go into state two and therefore automatically connecting
the set of state two resistors across the amplifiers 82, 83a
and 83b which cause the driver 43 to drive the operator 45 to
move the process drive 46 at a predetermined rapid speed in
the opposite direction. This is shown as state two in ~he
graph o Figure 33.

~l~S~7
In this state two, the summation of the error and the
rate change o~ the error will begin to be monitored and when
the polarity o~ this summation is again changed, the state
counter 156 will increment to state. three, thereby connecting
the set of state three resistors across the instrumentation
amplifier 82, ~he firs~ oparational amplifier 8~a and the
seccnd operational amplifier 83b thus causing this de~ice now
to act as a single-state four-mode controller identical to
that previously described. From the chart on Figure 34 it
can be seen that a great saving of time is achieved.
It should be understood that it is only the three-state
di~ferential input circuit 145, and in particular the three-
state error and rate amplifier circuit 146 therein, which is
changed to cause the controller to operate in these three
di~ferent states and achieve this great increase in speed.
The other circuits used in the single-state controller such as
the various corrective action circui~s, the valid range check
circui~, the scaling and meter protection circuit, ~he buffer-
scaler, the summing amplifier, th~ summing inte~rator, and the
absolute value circuits operate exactly the same as they did
before. It should be recognized that the polarity signal from
the absolute value circuit must also be connected to the three-
state error and ra~e amplifier circuit when using the correct-
ive action circuits shown in Figures 7 and 8.
As beore, the correction signal from the three-state
error and rate amplifier ci-rcuit is supplied to the driver 43,
7 S r

777
which in ~urn is supplied to the operator 45. As before the
operator may be any of several devices such as the two-
directional switched driver as shown in Figure 17, a reversible
AC synchronous motor shown in Figure 18, a reversible DC motor
as shown in Figure 19 or solenoids as shown in Figure 20.
Typically the process speed improvement device 128 is
a valve, while the second operator 127 is a solenoid, the
combination comprising a solenoid valve. The second driver
126 is any driver capable of converting a logic level signal
into a level capable of operating the process speed improve-
ment device, and in the case of operating the solenoid valve
might be one section of the quad 5 Amp DC driver as previously
listed.
.
When the process speed improvement device is used in
conjunction with our three-state four-mode process controller,
the state one signal ~rom the three-state error and rate
amplifier circui~ is connected to the second driver 126. In
this case, the process speed improvement device is operated
only when the three-sta~e four-mode process controller is in
its first s~ate.
If the process speed improvement device is used in
conjunction with the single-state process controller, ~he
signal to the second driver 1~6 would be ~ypically operated
either manually or by the automation device for a limited time
until the process correla~e signal approaches the desired
~ ~

~ 7~
value signal, thereby decreasing the time required to control
large changes in set point.
Another device which has proved particularly useful as
an operator in connec~ion with either the single-state or
three-state four-mode controllers of our invention is a DC
servo motor.. For the operation of such a motor, the correc-
t;on signal is supplied to a driver circuit whose function is
~o drive a DC servo motor in closed-loop operation so that the
motor speed and direction is a direct function of the voltage
and polarity of the correction signal. Details of such a
driver circuit are well Xnown in the art and can be found for
example by referring to the application note AN49 Incremental
Motion Servos, of PMI Motors, Division o Killmorgen Corpora-
tion, Syosset, New York.
-~ Thus, in addition to providing a single state control-
ler whi.ch uses the error difference and the rate of change of
the error difference to provide the best available controller
which conforms to past notions of con~roller theory and doesn't
overshoot~ by abandoning such past notions and intentionally
overshooting a set point we have developed a process con~rol-
ler which is much faster than those previously available.
- ~7 -